diff --git a/README.md b/README.md index 72e8331..56f145c 100644 --- a/README.md +++ b/README.md @@ -82,10 +82,22 @@ Choose **Publications**, select a paper or **Create an entry**, and fill in the title, year, authors (one per line), venue, paper URL, and publication type. Add research areas and related datasets where relevant. -Paste the abstract into **Abstract**. Use Markdown for formatting and LaTeX for math. -Full LaTeX documents and custom packages are not supported. +Use the published paper URL when available. Include full papers, not abstract-only +conference, poster, or demo entries. + +Paste the original abstract into **Abstract**. +Abstracts use **Markdown**, with **LaTeX equations** : + +- Text: `*italics*`, `**bold**`, and `[link text](https://example.org)`. +- Inline equations: `$x^2$` or `\(x^2\)`. +- Separate equations: `$$x^2$$` or `\[x^2\]`. + +Use `*text*` for italics, not `\textit{text}` outside an equation. +Full LaTeX documents and custom packages are not supported. Put extra details in **Additional notes**. +The archive shows 100 papers per page; search covers the entire archive. + ### Create or edit a personal page Save your member profile first, then choose **Personal Pages** and your diff --git a/assets/css/main.css b/assets/css/main.css index 70501bd..c58d9e9 100644 --- a/assets/css/main.css +++ b/assets/css/main.css @@ -1195,6 +1195,16 @@ a.chip { } [data-publication-controls][hidden] { display: none; } + +.publication-pagination { + display: flex; + flex-wrap: wrap; + align-items: center; + justify-content: center; + gap: 1rem; +} + +[data-publication-results] { scroll-margin-top: 6rem; } button { cursor: pointer; } button:disabled { cursor: default; opacity: 0.6; } diff --git a/assets/js/site.js b/assets/js/site.js index 4e92b64..f3f9d55 100644 --- a/assets/js/site.js +++ b/assets/js/site.js @@ -64,79 +64,175 @@ document.addEventListener("DOMContentLoaded", () => { } const searchInput = document.querySelector("[data-publication-search]"); - const cards = Array.from(document.querySelectorAll("[data-publication-card]")); - const emptyState = document.querySelector("[data-publication-empty]"); - const filterButtons = Array.from( - document.querySelectorAll("[data-publication-tag]"), - ); - if (searchInput) { + const archive = document.querySelector("[data-publication-archive]"); + const results = document.querySelector("[data-publication-results]"); + const paginations = [...document.querySelectorAll("[data-publication-pagination]")]; + const errorState = document.querySelector("[data-publication-error]"); + const emptyState = document.querySelector("[data-publication-empty]"); const controls = document.querySelector("[data-publication-controls]"); const researchSelect = document.querySelector("[data-publication-research]"); const resultCount = document.querySelector("[data-publication-count]"); const clearButton = document.querySelector("[data-publication-clear]"); + const filterButtons = [...document.querySelectorAll("[data-publication-tag]")]; const normalize = (text) => text.normalize("NFD").replace(/\p{M}/gu, "").replace(/[‘’]/g, "'").toLowerCase(); - const searchIndex = cards.map((card) => normalize(card.dataset.search || "")); + const pageSize = 100; + let records; + let indexRequest; let activeTag = "all"; - - const applyFilter = (updateURL = true) => { + let currentPage = Number(archive.dataset.publicationPage); + let revision = 0; + let searchTimer; + + // The initial page is server-rendered. Fetch the searchable metadata once, + // only when filters (including a shared search URL) are used. + const loadIndex = async () => { + if (!indexRequest) { + indexRequest = fetch(archive.dataset.publicationIndex) + .then((response) => { + if (!response.ok) throw new Error("Publication index unavailable"); + return response.json(); + }) + .then((data) => { + records = data.map((record) => ({ ...record, search: normalize(record.search) })); + return records; + }) + .catch((error) => { indexRequest = null; throw error; }); + } + return indexRequest; + }; + const element = (tag, text, className) => { + const node = document.createElement(tag); + if (text !== undefined) node.textContent = text; + if (className) node.className = className; + return node; + }; + const cardFor = (record) => { + const card = element("article", undefined, "list-card publication-card"); + card.dataset.publicationCard = ""; + const title = element("h2"); + const link = element("a", record.title); + link.href = record.url; + title.append(link); + card.append(element("p", record.authors.join(" / "), "muted"), title, + element("p", `${record.venue}, ${record.year} · ${record.type}`)); + if (record.tags.length) { + const tags = element("div", undefined, "chip-row"); + record.tags.forEach((tag) => tags.append(element("span", tag, "chip chip-muted"))); + card.append(tags); + } + return card; + }; + const filterURL = (page = currentPage) => { + const url = new URL(archive.dataset.publicationBase, location.href); + for (const [key, value] of [["q", searchInput.value.trim()], ["tag", activeTag], ["research", researchSelect?.value]]) { + if (value && value !== "all") url.searchParams.set(key, value); + } + if (page > 1) url.searchParams.set("page", page); + return url; + }; + const renderPagination = (pageCount) => { + paginations.forEach((nav) => { + nav.replaceChildren(); + const addLink = (label, page, relation) => { + const link = element("a", label, "button button-secondary"); + link.href = filterURL(page); + link.rel = relation; + link.addEventListener("click", (event) => { + if (event.ctrlKey || event.metaKey || event.shiftKey || event.altKey || event.button !== 0) return; + event.preventDefault(); + clearTimeout(searchTimer); + currentPage = page; + history.pushState(null, "", filterURL()); + applyFilter(false).then(() => { + results.focus({ preventScroll: true }); + results.scrollIntoView({ block: "start" }); + }); + }); + nav.append(link); + }; + if (currentPage > 1) addLink("Previous", currentPage - 1, "prev"); + nav.append(element("span", `Page ${currentPage} of ${pageCount}`)); + if (currentPage < pageCount) addLink("Next", currentPage + 1, "next"); + }); + }; + const applyFilter = async (updateURL = true) => { + const requestRevision = ++revision; + if (!records) resultCount.textContent = "Loading publication search…"; + try { await loadIndex(); } catch { + if (requestRevision === revision) { + errorState.hidden = false; + resultCount.textContent = "Browse publications using the page links below."; + } + return; + } + if (requestRevision !== revision) return; + errorState.hidden = true; const terms = normalize(searchInput.value.trim()).split(/\s+/).filter(Boolean); const research = researchSelect?.value || "all"; - let visible = 0; - cards.forEach((card, index) => { - const tags = (card.dataset.tags || "").split("|"); - const areas = (card.dataset.research || "").split("|"); - const match = terms.every((term) => searchIndex[index].includes(term)) && - (activeTag === "all" || tags.includes(activeTag)) && - (research === "all" || areas.includes(research)); - card.hidden = !match; - if (match) visible += 1; - }); + const matches = records.filter((record) => terms.every((term) => record.search.includes(term)) && + (activeTag === "all" || record.tags.includes(activeTag)) && + (research === "all" || record.research.includes(research))); + const pageCount = Math.max(1, Math.ceil(matches.length / pageSize)); + currentPage = Math.min(Math.max(1, currentPage), pageCount); + const start = (currentPage - 1) * pageSize; + results.replaceChildren(...matches.slice(start, start + pageSize).map(cardFor)); filterButtons.forEach((button) => { const selected = button.dataset.publicationTag === activeTag; button.classList.toggle("is-active", selected); button.setAttribute("aria-pressed", String(selected)); }); - if (emptyState) emptyState.hidden = visible !== 0; - if (resultCount) resultCount.textContent = `${visible} of ${cards.length} publications`; - if (clearButton) clearButton.disabled = !searchInput.value && activeTag === "all" && research === "all"; - if (updateURL) { - const url = new URL(location.href); - url.searchParams.delete("author"); - for (const [key, value] of [["q", searchInput.value.trim()], ["tag", activeTag], ["research", research]]) { - if (value && value !== "all") url.searchParams.set(key, value); - else url.searchParams.delete(key); - } - history.replaceState(null, "", url); - } + emptyState.hidden = matches.length !== 0; + resultCount.textContent = matches.length ? + `Showing ${start + 1}–${Math.min(start + pageSize, matches.length)} of ${matches.length} publications` : + "0 publications"; + renderPagination(pageCount); + clearButton.disabled = !searchInput.value && activeTag === "all" && research === "all"; + if (updateURL) history.replaceState(null, "", filterURL()); }; const restoreFilter = () => { - const params = new URL(location.href).searchParams; + clearTimeout(searchTimer); + ++revision; + const url = new URL(location.href); + const params = url.searchParams; searchInput.value = params.get("q") || params.get("author") || ""; const tag = params.get("tag"); activeTag = filterButtons.some((button) => button.dataset.publicationTag === tag) ? tag : "all"; - if (researchSelect) { - const area = params.get("research"); - researchSelect.value = [...researchSelect.options].some((option) => option.value === area) ? area : "all"; - } - applyFilter(false); + const area = params.get("research"); + researchSelect.value = [...researchSelect.options].some((option) => option.value === area) ? area : "all"; + const page = params.get("page") || url.pathname.match(/\/page\/(\d+)\//)?.[1] || "1"; + currentPage = /^[1-9]\d*$/.test(page) && Number.isSafeInteger(Number(page)) ? Number(page) : 1; + const filtered = searchInput.value || activeTag !== "all" || researchSelect.value !== "all"; + if (records || filtered || params.has("page")) applyFilter(false); + clearButton.disabled = !filtered; }; - searchInput.addEventListener("input", () => applyFilter()); - researchSelect?.addEventListener("change", () => applyFilter()); + const changeFilter = () => { + clearTimeout(searchTimer); + currentPage = 1; + applyFilter(); + }; + searchInput.addEventListener("input", () => { + clearTimeout(searchTimer); + ++revision; + currentPage = 1; + searchTimer = setTimeout(() => applyFilter(), 150); + }); + researchSelect.addEventListener("change", changeFilter); filterButtons.forEach((button) => button.addEventListener("click", () => { activeTag = button.dataset.publicationTag || "all"; - applyFilter(); + changeFilter(); })); - clearButton?.addEventListener("click", () => { + clearButton.addEventListener("click", () => { searchInput.value = ""; activeTag = "all"; - if (researchSelect) researchSelect.value = "all"; - applyFilter(); + researchSelect.value = "all"; + changeFilter(); searchInput.focus(); }); + document.querySelector("[data-publication-retry]").addEventListener("click", () => applyFilter()); window.addEventListener("popstate", restoreFilter); restoreFilter(); - if (controls) controls.hidden = false; + controls.hidden = false; } const msrgGame = document.querySelector("[data-msrg-game]"); diff --git a/content/people/alex-cheung.md b/content/people/alex-cheung.md index 82261b6..bdd00e3 100644 --- a/content/people/alex-cheung.md +++ b/content/people/alex-cheung.md @@ -7,4 +7,5 @@ start_date = "2004-09" end_date = "2009-06" location = "MSRG alumnus" research = ["data-management"] +author_names = ["Alex Cheung", "Alex King Yeung Cheung"] +++ diff --git a/content/people/alexander-erben.md b/content/people/alexander-erben.md index 7ed4072..a31cd86 100644 --- a/content/people/alexander-erben.md +++ b/content/people/alexander-erben.md @@ -10,4 +10,5 @@ research = ["distributed-machine-learning"] orcid = "https://orcid.org/0000-0002-0153-7251" linkedin = "https://www.linkedin.com/in/alexandererben/" end_date = "2024" +author_names = ["Alexander Erben", "Alexander Isenko"] +++ diff --git a/content/people/herbert-woisetschlaeger.md b/content/people/herbert-woisetschlaeger.md index 60b7469..8b19dc0 100644 --- a/content/people/herbert-woisetschlaeger.md +++ b/content/people/herbert-woisetschlaeger.md @@ -10,4 +10,5 @@ interests = ["Federated Learning", "Efficient Machine Learning"] portrait = "/images/people/herbert-woisetschlaeger.jpg" homepage = "https://research.ibm.com/people/herbert-woisetschlaeger" linkedin = "https://www.linkedin.com/in/hwoisetschlaeger/" +author_names = ["Herbert Woisetschläger", "Herbert Woisetschlaeger"] +++ diff --git a/content/people/jose-rivera.md b/content/people/jose-rivera.md index f378128..bd89b23 100644 --- a/content/people/jose-rivera.md +++ b/content/people/jose-rivera.md @@ -7,4 +7,5 @@ status = "alumni" location = "MSRG alumnus" research = ["data-management"] homepage = "https://www.cs.cit.tum.de/dis/alumni/dr-jose-adan-rivera-acevedo/" +author_names = ["José Rivera", "Jose Rivera"] +++ diff --git a/content/people/lixia-chen.md b/content/people/lixia-chen.md index 4aea772..0c5fc26 100644 --- a/content/people/lixia-chen.md +++ b/content/people/lixia-chen.md @@ -6,4 +6,5 @@ status = "alumni" location = "MSRG alumna" research = ["data-management"] orcid = "https://orcid.org/0000-0001-7943-3623" +author_names = ["Lixia Chen", "Li-Xia Chen"] +++ diff --git a/content/people/michalis-bachras.md b/content/people/michalis-bachras.md index 8db8009..689255b 100644 --- a/content/people/michalis-bachras.md +++ b/content/people/michalis-bachras.md @@ -11,4 +11,5 @@ portrait = "/images/people/michalis-bachras.jpg" linkedin = "https://www.linkedin.com/in/michalis-bachras/" email = "michalis.bachras@mail.utoronto.ca" # Contact address supplied in the maintainer's group-member list, 2026-09-09. +author_names = ["Michalis Bachras", "Michail Bachras"] +++ diff --git a/content/people/mohammadreza-najafi.md b/content/people/mohammadreza-najafi.md index 4595679..f69b2bf 100644 --- a/content/people/mohammadreza-najafi.md +++ b/content/people/mohammadreza-najafi.md @@ -8,4 +8,5 @@ location = "MSRG alumnus" research = ["data-management"] orcid = "https://orcid.org/0000-0002-0629-7955" homepage = "https://www.cs.cit.tum.de/dis/alumni/dr-mohammadreza-najafi/" +author_names = ["Mohammadreza Najafi", "Mohammedreza Najafi"] +++ diff --git a/content/people/victor-del-razo.md b/content/people/victor-del-razo.md index f12c0be..7bdfcaf 100644 --- a/content/people/victor-del-razo.md +++ b/content/people/victor-del-razo.md @@ -9,4 +9,5 @@ orcid = "https://orcid.org/0000-0002-6807-9961" linkedin = "https://www.linkedin.com/in/victordelrazo/" start_date = "2012" end_date = "2016" +author_names = ["Victor Del Razo", "Victor del Razo"] +++ diff --git a/content/people/yuqiu-zhang.md b/content/people/yuqiu-zhang.md index a8ee757..1170e0e 100644 --- a/content/people/yuqiu-zhang.md +++ b/content/people/yuqiu-zhang.md @@ -12,4 +12,5 @@ interests = ["Cloud Resource Management", "Serverless Computing", "Stream Proces portrait = "/images/people/yuqiu-zhang.jpg" homepage = "https://qzhang.ca/" linkedin = "https://www.linkedin.com/in/yuqiu-quincy-zhang/" +author_names = ["Quincy Yuqiu Zhang", "Yuqiu Zhang"] +++ diff --git a/content/publications/2ps-high-quality-edge-partitioning-with-two-phase-streaming.md b/content/publications/2ps-high-quality-edge-partitioning-with-two-phase-streaming.md new file mode 100644 index 0000000..cfce69e --- /dev/null +++ b/content/publications/2ps-high-quality-edge-partitioning-with-two-phase-streaming.md @@ -0,0 +1,10 @@ ++++ +title = "2PS: High-Quality Edge Partitioning with Two-Phase Streaming" +year = 2020 +authors = ["Ruben Mayer", "Kamil Orujzade", "Hans-Arno Jacobsen"] +venue = "arXiv" +publication_type = "ArXiv Preprint" +research = ["data-management"] +external_url = "https://arxiv.org/abs/2001.07086" +abstract = "Graph partitioning is an important preprocessing step to distributed graph processing. In edge partitioning, the edge set of a given graph is split into $k$ equally-sized partitions, such that the replication of vertices across partitions is minimized. Streaming is a viable approach to partition graphs that exceed the memory capacities of a single server. The graph is ingested as a stream of edges, and one edge at a time is immediately and irrevocably assigned to a partition based on a scoring function. However, streaming partitioning suffers from the uninformed assignment problem: At the time of partitioning early edges in the stream, there is no information available about the rest of the edges. As a consequence, edge assignments are often driven by balancing considerations, and the achieved replication factor is comparably high. In this paper, we propose 2PS, a novel two-phase streaming algorithm for high-quality edge partitioning. In the first phase, vertices are separated into clusters by a lightweight streaming clustering algorithm. In the second phase, the graph is re-streamed and edge partitioning is performed while taking into account the clustering of the vertices from the first phase. Our evaluations show that 2PS can achieve a replication factor that is comparable to heavy-weight random access partitioners while inducing orders of magnitude lower memory overhead." ++++ diff --git a/content/publications/_index.md b/content/publications/_index.md index f412b60..22ffc75 100644 --- a/content/publications/_index.md +++ b/content/publications/_index.md @@ -1,5 +1,6 @@ +++ title = "Publications" layout = "archive" +outputs = ["HTML", "JSON"] summary = "Research papers, conference proceedings, and technical reports from MSRG spanning data management, distributed machine learning, and quantum computing systems." +++ diff --git a/content/publications/a-bigbench-implementation-in-the-hadoop-ecosystem.md b/content/publications/a-bigbench-implementation-in-the-hadoop-ecosystem.md new file mode 100644 index 0000000..2c0a657 --- /dev/null +++ b/content/publications/a-bigbench-implementation-in-the-hadoop-ecosystem.md @@ -0,0 +1,9 @@ ++++ +title = "A BigBench Implementation in the Hadoop Ecosystem" +year = 2014 +authors = ["Badrul Chowdhury", "Tilmann Rabl", "Pooya Saadatpanah", "Jiang Du", "Hans-Arno Jacobsen"] +venue = "Lecture Notes in Computer Science; Advancing Big Data Benchmarks" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1007/978-3-319-10596-3_1" ++++ diff --git a/content/publications/a-blockchain-based-privacy-preserving-charging-station-reservation-and-payment-scheme-for-electric-vehicles.md b/content/publications/a-blockchain-based-privacy-preserving-charging-station-reservation-and-payment-scheme-for-electric-vehicles.md new file mode 100644 index 0000000..96c28c9 --- /dev/null +++ b/content/publications/a-blockchain-based-privacy-preserving-charging-station-reservation-and-payment-scheme-for-electric-vehicles.md @@ -0,0 +1,10 @@ ++++ +title = "A Blockchain-Based Privacy-Preserving Charging Station Reservation and Payment Scheme for Electric Vehicles" +year = 2025 +authors = ["Syed Muhammad Danish", "Muhammad Muneem Shabir", "Kaiwen Zhang", "Hans-Arno Jacobsen", "Syed Ali Hassan"] +venue = "Distributed Ledger Technologies: Research and Practice" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.1145/3696428" +abstract = "EV charging infrastructures traditionally rely on untrusted centralized infrastructures that pose several privacy and security threats to EVs’ personal information. Targeted advertisements, privacy leaks and selling data to third parties are among the threats to privacy and security. By utilizing blockchain-based solutions, recent work address the security and privacy problems associated with EV charging protocols. Most of them are geared toward maintaining EV anonymity rather than preserving end-to-end privacy. As EV owners’ charging histories and payment information are associated with their wallet addresses on the blockchain, any threat of linkability of these blockchain addresses to physical identities can pose a serious risk to their privacy. In this paper, we propose a ring signature based privacy-preserving end-to-end charging station (CS) reservation and payment protocol, which provides EV owners with the ability to reserve and pay for a charging slot privately without sharing private information or exposing their identity or addresses at CS locations. Additionally, we provide EV owners with a decentralized charging slot information verification protocol with the help of secure multiparty computation (SMC), which allows them to verify available slots. A dispute resolution mechanism is also proposed that handles disputes between EVs and CSs and penalizes them accordingly by utilizing trusted execution environment (TEE). Results show that the proposed protocol ensures end-to-end EV owners’ privacy with low blockchain transaction and computation overhead." ++++ diff --git a/content/publications/a-blockchain-based-privacy-preserving-intelligent-charging-station-selection-for-electric-vehicles.md b/content/publications/a-blockchain-based-privacy-preserving-intelligent-charging-station-selection-for-electric-vehicles.md new file mode 100644 index 0000000..3a7b42d --- /dev/null +++ b/content/publications/a-blockchain-based-privacy-preserving-intelligent-charging-station-selection-for-electric-vehicles.md @@ -0,0 +1,10 @@ ++++ +title = "A Blockchain-Based Privacy-Preserving Intelligent Charging Station Selection for Electric Vehicles" +year = 2020 +authors = ["Syed Muhammad Danish", "Kaiwen Zhang", "Hans-Arno Jacobsen"] +venue = "2020 IEEE International Conference on Blockchain and Cryptocurrency (ICBC)" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icbc48266.2020.9169419" +abstract = "The untrusted centralized nature of energy markets and electric vehicle (EV) charging infrastructures result in several privacy and security threats to the private information of EV users. These security and privacy threats include targeted advertisements, privacy leakage, selling data to third party, etc. In this work, we propose a blockchain-based privacy-preserving intelligent charging station (CS) selection for EVs to ensure the security and privacy of the EV users and availability of the CSs. We introduce a blockchain-based framework to implement secure charging services and trusted reservation for EVs through the execution of smart contracts. We also formulate the problem of privacy-preserving intelligent CS selection and propose a mechanism for EVs to select the optimal CS locally based on dynamic requirements. Finally, we present an example scenario of our proposed framework." ++++ diff --git a/content/publications/a-comprehensive-feature-study-for-appliance-recognition-on-high-frequency-energy-data.md b/content/publications/a-comprehensive-feature-study-for-appliance-recognition-on-high-frequency-energy-data.md new file mode 100644 index 0000000..8963e73 --- /dev/null +++ b/content/publications/a-comprehensive-feature-study-for-appliance-recognition-on-high-frequency-energy-data.md @@ -0,0 +1,10 @@ ++++ +title = "A Comprehensive Feature Study for Appliance Recognition on High Frequency Energy Data" +year = 2017 +authors = ["Matthias Kahl", "Anwar Ul Haq", "Thomas Kriechbaumer", "Hans-Arno Jacobsen"] +venue = "Proceedings of the Eighth International Conference on Future Energy Systems" +publication_type = "Conference Paper" +research = [] +external_url = "https://doi.org/10.1145/3077839.3077845" +abstract = "Awareness about the energy consumption of appliances can help to save energy in households. Non-intrusive Load Monitoring (NILM) is a feasible approach to provide consumption feedback at appliance level. In this paper, we evaluate a broad set of features for electrical appliance recognition, extracted from high frequency start-up events. These evaluations were applied on several existing high frequency energy datasets. To examine clean signatures, we ran all experiments on two datasets that are based on isolated appliance events; more realistic results were retrieved from two real household datasets. Our feature set consists of 36 signatures from related work including novel approaches, and from other research fields. The results of this work include a stand-alone feature ranking, promising feature combinations for appliance recognition in general and appliance-wise performances." ++++ diff --git a/content/publications/a-comprehensive-study-on-benchmarking-permissioned-blockchains.md b/content/publications/a-comprehensive-study-on-benchmarking-permissioned-blockchains.md new file mode 100644 index 0000000..719f1f4 --- /dev/null +++ b/content/publications/a-comprehensive-study-on-benchmarking-permissioned-blockchains.md @@ -0,0 +1,9 @@ ++++ +title = "A Comprehensive Study on Benchmarking Permissioned Blockchains" +year = 2024 +authors = ["Jeeta Ann Chacko", "Ruben Mayer", "Alan D. Fekete", "Vincent Gramoli", "Hans-Arno Jacobsen"] +venue = "Lecture Notes in Computer Science; Performance Evaluation and Benchmarking" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1007/978-3-031-68031-1_2" ++++ diff --git a/content/publications/a-comprehensive-survey-of-machine-unlearning-techniques-for-large-language-models.md b/content/publications/a-comprehensive-survey-of-machine-unlearning-techniques-for-large-language-models.md new file mode 100644 index 0000000..fadb062 --- /dev/null +++ b/content/publications/a-comprehensive-survey-of-machine-unlearning-techniques-for-large-language-models.md @@ -0,0 +1,10 @@ ++++ +title = "A Comprehensive Survey of Machine Unlearning Techniques for Large Language Models" +year = 2025 +authors = ["Jiahui Geng", "Qing Li", "Herbert Woisetschlaeger", "Zongxiong Chen", "Fengyu Cai", "Yuxia Wang", "Preslav Nakov", "Hans-Arno Jacobsen", "Fakhri Karray"] +venue = "arXiv" +publication_type = "ArXiv Preprint" +research = ["distributed-machine-learning"] +external_url = "https://arxiv.org/abs/2503.01854" +abstract = "This study investigates the machine unlearning techniques within the context of large language models (LLMs), referred to as *LLM unlearning*. LLM unlearning offers a principled approach to removing the influence of undesirable data (e.g., sensitive or illegal information) from LLMs, while preserving their overall utility without requiring full retraining. Despite growing research interest, there is no comprehensive survey that systematically organizes existing work and distills key insights; here, we aim to bridge this gap. We begin by introducing the definition and the paradigms of LLM unlearning, followed by a comprehensive taxonomy of existing unlearning studies. Next, we categorize current unlearning approaches, summarizing their strengths and limitations. Additionally, we review evaluation metrics and benchmarks, providing a structured overview of current assessment methodologies. Finally, we outline promising directions for future research, highlighting key challenges and opportunities in the field." ++++ diff --git a/content/publications/a-crowdsourcing-approach-for-the-inference-of-distribution-grids.md b/content/publications/a-crowdsourcing-approach-for-the-inference-of-distribution-grids.md new file mode 100644 index 0000000..ac12e8e --- /dev/null +++ b/content/publications/a-crowdsourcing-approach-for-the-inference-of-distribution-grids.md @@ -0,0 +1,10 @@ ++++ +title = "A Crowdsourcing Approach for the Inference of Distribution Grids" +year = 2018 +authors = ["Pezhman Nasirifard", "José Rivera", "Qunjie Zhou", "Klaus Bernd Schreiber", "Hans-Arno Jacobsen"] +venue = "Proceedings of the Ninth International Conference on Future Energy Systems" +publication_type = "Conference Paper" +research = ["distributed-machine-learning"] +external_url = "https://doi.org/10.1145/3208903.3208927" +abstract = "Maintaining a complete and up-to-date model of the distribution grid is a challenging task, and the scarcity of open models represents a significant bottleneck for researchers in this area. In this work, we address these challenges by introducing a crowdsourcing framework for the collection of open data on distribution grid devices and an algorithm to infer the topological model of the distribution grids. We use the crowd and smartphones to collect an image and the geographical position of power distribution grid devices. Since power distribution lines are usually underground and cannot be mapped, we use spatial data analytics on the collected data in combination with other open data sources to infer the topology of the distribution grid. This paper describes and evaluates our crowdsourcing and inference approach. To evaluate our approach, we organized and conducted a crowdsourcing campaign to map and infer a sizeable district in Munich, Germany. The results are compared with the ground truth of the distribution system operator. Our field experiments show that using the crowd to recognize power distribution elements, a precision of up to 82% and a recall of up to 65% can be obtained. The numerical evaluation of our inference algorithm demonstrates that the model we inferred based on the acquired official DSO grid dataset achieves a power length accuracy of 88% compared to the ground truth. These results confirm our approach as a practical method to infer real power distribution grid models." ++++ diff --git a/content/publications/a-distributed-anytime-algorithm-for-network-utility-maximization-with-application-to-real-time-ev-charging-control.md b/content/publications/a-distributed-anytime-algorithm-for-network-utility-maximization-with-application-to-real-time-ev-charging-control.md new file mode 100644 index 0000000..611feca --- /dev/null +++ b/content/publications/a-distributed-anytime-algorithm-for-network-utility-maximization-with-application-to-real-time-ev-charging-control.md @@ -0,0 +1,10 @@ ++++ +title = "A Distributed anytime algorithm for Network Utility Maximization with application to real-time EV charging control" +year = 2014 +authors = ["José Rivera", "Hans-Arno Jacobsen"] +venue = "53rd IEEE Conference on Decision and Control" +publication_type = "Conference Paper" +research = [] +external_url = "https://doi.org/10.1109/cdc.2014.7039503" +abstract = "The control of Electric Vehicle (EV) charging to take full advantage of the available distribution network infrastructure represents a cornerstone for large-scale EV adoption and the reduction of greenhouse gas emissions. In this paper, we propose a novel distributed anytime algorithm to solve the Network Utility Maximization (NUM) problem with application to real-time EV charging control. We analyze its convergence conditions for synchronous and asynchronous execution. Beyond this, we evaluate our approach using real data and show its advantages against the standard dual decomposition approach. The control scheme in our approach is based on the notion of dynamic budgets defined by the protection devices and allocated to each EV charger. Given the system's current state, we solve EV charging as a NUM problem in a distributed manner and obtain closed form expressions for computations performed by EV chargers and protection devices. To cope with large EV numbers, their spatial distribution, and the highly dynamic state changes of the power grid, our approach allows for distributed computation capable of yielding feasible, albeit suboptimal, control values at any time." ++++ diff --git a/content/publications/a-distributed-anytime-algorithm-for-real-time-ev-charging-congestion-control.md b/content/publications/a-distributed-anytime-algorithm-for-real-time-ev-charging-congestion-control.md new file mode 100644 index 0000000..686a9a8 --- /dev/null +++ b/content/publications/a-distributed-anytime-algorithm-for-real-time-ev-charging-congestion-control.md @@ -0,0 +1,10 @@ ++++ +title = "A Distributed Anytime Algorithm for Real-Time EV Charging Congestion Control" +year = 2015 +authors = ["José Rivera", "Christoph Goebel", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 2015 ACM Sixth International Conference on Future Energy Systems" +publication_type = "Conference Paper" +research = [] +external_url = "https://doi.org/10.1145/2768510.2768544" +abstract = "A massive introduction of Electric Vehicles (EVs) will cause considerable load increase, and without proper control, can lead to congestion problems in the distribution of power. EV charging congestion control is the ability to control the EVs' charging rate to avoid the overloading of distribution grid elements while optimizing the use of the available infrastructure. We propose the implementation of a distributed anytime algorithm for real-time congestion control of EV charging in distribution feeders. The problem is formulated as a network utility maximization problem. An iterative distributed solution algorithm with feasible iterates is investigated. This paper shows the formulation of the proposed solution approach and also the formulation of the state-of-the-art dual decomposition solution. The resulting algorithms for the former approaches are evaluated under static and dynamic conditions. The results demonstrate that the proposed algorithm remains stable and retains its anytime property under dynamic conditions. Compared to the state-of-the-art solution, the proposed algorithm offers improved scalability and reliability for EV charging congestion control." ++++ diff --git a/content/publications/a-distributed-framework-for-reliable-and-efficient-service-choreographies.md b/content/publications/a-distributed-framework-for-reliable-and-efficient-service-choreographies.md new file mode 100644 index 0000000..d781976 --- /dev/null +++ b/content/publications/a-distributed-framework-for-reliable-and-efficient-service-choreographies.md @@ -0,0 +1,10 @@ ++++ +title = "A distributed framework for reliable and efficient service choreographies" +year = 2011 +authors = ["Young Yoon", "Chunyang Ye", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 20th international conference on World wide web" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/1963405.1963515" +abstract = "In service-oriented architectures (SOA), independently developed Web services can be dynamically composed. However, the composition is prone to producing semantically conflicting interactions among the services. For example, in an interdepartmental business collaboration through Web services, the decision by the marketing department to clear out the inventory might be inconsistent with the decision by the operations department to increase production. Resolving semantic conflicts is challenging especially when services are loosely coupled and their interactions are not carefully governed. To address this problem, we propose a novel distributed service choreography framework. We deploy safety constraints to prevent conflicting behavior and enforce reliable and efficient service interactions via federated publish/subscribe messaging, along with strategic placement of distributed choreography agents and coordinators to minimize runtime overhead. Experimental results show that our framework prevents semantic conflicts with negligible overhead and scales better than a centralized approach by up to 60%." ++++ diff --git a/content/publications/a-distributed-service-oriented-architecture-for-business-process-execution.md b/content/publications/a-distributed-service-oriented-architecture-for-business-process-execution.md new file mode 100644 index 0000000..fa0979b --- /dev/null +++ b/content/publications/a-distributed-service-oriented-architecture-for-business-process-execution.md @@ -0,0 +1,10 @@ ++++ +title = "A distributed service-oriented architecture for business process execution" +year = 2010 +authors = ["Guoli Li", "Vinod Muthusamy", "Hans-Arno Jacobsen"] +venue = "ACM Transactions on the Web" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.1145/1658373.1658375" +abstract = "The Business Process Execution Language (BPEL) standardizes the development of composite enterprise applications that make use of software components exposed as Web services. BPEL processes are currently executed by a centralized orchestration engine, in which issues such as scalability, platform heterogeneity, and division across administrative domains can be difficult to manage. We propose a distributed agent-based orchestration engine in which several lightweight agents execute a portion of the original business process and collaborate in order to execute the complete process. The complete set of standard BPEL activities are supported, and the transformations of several BPEL activities to the agent-based architecture are described. Evaluations of an implementation of this architecture demonstrate that agent-based execution scales better than a non-distributed approach, with at least 70% and 120% improvements in process execution time, and throughput, respectively, even with a large number of concurrent process instances. In addition, the distributed architecture successfully executes large processes that are shown to be infeasible to execute with a nondistributed engine." ++++ diff --git a/content/publications/a-framework-for-control-and-co-simulation-in-distribution-networks-applied-to-electric-vehicle-charging-with-vehicle-originating-signals.md b/content/publications/a-framework-for-control-and-co-simulation-in-distribution-networks-applied-to-electric-vehicle-charging-with-vehicle-originating-signals.md new file mode 100644 index 0000000..7a9eb9b --- /dev/null +++ b/content/publications/a-framework-for-control-and-co-simulation-in-distribution-networks-applied-to-electric-vehicle-charging-with-vehicle-originating-signals.md @@ -0,0 +1,10 @@ ++++ +title = "A framework for control and co-simulation in distribution networks applied to electric vehicle charging with Vehicle-Originating-Signals" +year = 2016 +authors = ["Hamidreza Mirtaheri", "Gianfranco Chicco", "Victor del Razo", "Hans-Arno Jacobsen"] +venue = "2016 IEEE International Energy Conference (ENERGYCON)" +publication_type = "Conference Paper" +research = [] +external_url = "https://doi.org/10.1109/energycon.2016.7513941" +abstract = "The increasing share of renewable energy sources and distributed energy generation (DER) in the energy mix brings a number of challenges to the power system, particularly to the distribution network (DN). New control methods and algorithms are being studied to address these challenges, but the effects of such algorithms on actual DNs are often difficult to validate. Co-simulation, considering the advanced state of existing DN simulation tools, is a potential solution but usually requires some degree of expertise. In this work, we introduce a framework for implementing control algorithms for DNs assisted by co-simulation. This framework uses the commercial simulator PowerFactory but does not require a high tool-specific expertise and can be integrated into other simulation tools. Then we apply this framework to an electric vehicle charging use case and show the results and benefits. Furthermore we extend an existing charging control algorithm to support local voltage control." ++++ diff --git a/content/publications/a-generalized-algorithm-for-publish-subscribe-overlay-design-and-its-fast-implementation.md b/content/publications/a-generalized-algorithm-for-publish-subscribe-overlay-design-and-its-fast-implementation.md new file mode 100644 index 0000000..5bba5b1 --- /dev/null +++ b/content/publications/a-generalized-algorithm-for-publish-subscribe-overlay-design-and-its-fast-implementation.md @@ -0,0 +1,9 @@ ++++ +title = "A Generalized Algorithm for Publish/Subscribe Overlay Design and Its Fast Implementation" +year = 2012 +authors = ["Chen Chen", "Roman Vitenberg", "Hans-Arno Jacobsen"] +venue = "Lecture Notes in Computer Science; Distributed Computing" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1007/978-3-642-33651-5_6" ++++ diff --git a/content/publications/a-hybrid-b-tree-as-solution-for-in-memory-indexing-on-cpu-gpu-heterogeneous-computing-platforms.md b/content/publications/a-hybrid-b-tree-as-solution-for-in-memory-indexing-on-cpu-gpu-heterogeneous-computing-platforms.md new file mode 100644 index 0000000..bb72027 --- /dev/null +++ b/content/publications/a-hybrid-b-tree-as-solution-for-in-memory-indexing-on-cpu-gpu-heterogeneous-computing-platforms.md @@ -0,0 +1,10 @@ ++++ +title = "A Hybrid B+-tree as Solution for In-Memory Indexing on CPU-GPU Heterogeneous Computing Platforms" +year = 2016 +authors = ["Amirhesam Shahvarani", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 2016 International Conference on Management of Data" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/2882903.2882918" +abstract = "An in-memory indexing tree is a critical component of many databases. Modern many-core processors, such as GPUs, are offering tremendous amounts of computing power making them an attractive choice for accelerating indexing. However, the memory available to the accelerating co-processor is rather limited and expensive in comparison to the memory available to the CPU. This drawback is a barrier to exploit the computing power of co-processors for arbitrarily large index trees. In this paper, we propose a novel design for a B+-tree based on the heterogeneous computing platform and the hybrid memory architecture found in GPUs. We propose a hybrid CPU-GPU B+-tree, \"HB+-tree,\" which targets high search throughput use cases. Unique to our design is the joint and simultaneous use of computing and memory resources of CPU-GPU systems. Our experiments show that our HB+-tree can perform up to 240 million index queries per second, which is 2.4X higher than our CPU-optimized solution." ++++ diff --git a/content/publications/a-memory-bandwidth-efficient-hybrid-radix-sort-on-gpus.md b/content/publications/a-memory-bandwidth-efficient-hybrid-radix-sort-on-gpus.md new file mode 100644 index 0000000..9a8f3a7 --- /dev/null +++ b/content/publications/a-memory-bandwidth-efficient-hybrid-radix-sort-on-gpus.md @@ -0,0 +1,10 @@ ++++ +title = "A Memory Bandwidth-Efficient Hybrid Radix Sort on GPUs" +year = 2017 +authors = ["Elias Stehle", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 2017 ACM International Conference on Management of Data" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/3035918.3064043" +abstract = "Sorting is at the core of many database operations, such as index creation, sort-merge joins, and user-requested output sorting. As GPUs are emerging as a promising platform to accelerate various operations, sorting on GPUs becomes a viable endeavour. Over the past few years, several improvements have been proposed for sorting on GPUs, leading to the first radix sort implementations that achieve a sorting rate of over one billion 32-bit keys per second. Yet, state-of-the-art approaches are heavily memory bandwidth-bound, as they require substantially more memory transfers than their CPU-based counterparts. Our work proposes a novel approach that almost halves the amount of memory transfers and, therefore, considerably lifts the memory bandwidth limitation. Being able to sort two gigabytes of eight-byte records in as little as 50 milliseconds, our approach achieves a 2.32-fold improvement over the state-of-the-art GPU-based radix sort for uniform distributions, sustaining a minimum speed-up of no less than a factor of 1.66 for skewed distributions. To address inputs that either do not reside on the GPU or exceed the available device memory, we build on our efficient GPU sorting approach with a pipelined heterogeneous sorting algorithm that mitigates the overhead associated with PCIe data transfers. Comparing the end-to-end sorting performance to the state-of-the-art CPU-based radix sort running 16 threads, our heterogeneous approach achieves a 2.06-fold and a 1.53-fold improvement for sorting 64 GB key-value pairs with a skewed and a uniform distribution, respectively." ++++ diff --git a/content/publications/a-scalable-circular-pipeline-design-for-multi-way-stream-joins-in-hardware.md b/content/publications/a-scalable-circular-pipeline-design-for-multi-way-stream-joins-in-hardware.md new file mode 100644 index 0000000..92e5091 --- /dev/null +++ b/content/publications/a-scalable-circular-pipeline-design-for-multi-way-stream-joins-in-hardware.md @@ -0,0 +1,10 @@ ++++ +title = "A Scalable Circular Pipeline Design for Multi-Way Stream Joins in Hardware" +year = 2018 +authors = ["Mohammadreza Najafi", "Mohammad Sadoghi", "Hans-Arno Jacobsen"] +venue = "2018 IEEE 34th International Conference on Data Engineering (ICDE)" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icde.2018.00130" +abstract = "Efficient real-time analytics are an integral part of a growing number of data management applications such as computational targeted advertising, algorithmic trading, and Internet of Things. In this paper, we primarily focus on accelerating stream joins, arguably one of the most commonly used and resource-intensive operators in stream processing. We propose a scalable circular pipeline design (Circular-MJ) in hardware to orchestrate multi-way join while minimizing data flow disruption. In this circular design, each new tuple (given its origin stream) starts its processing from a specific join core and passes through all respective join cores in a pipeline sequence to produce final results. We further present a novel two-stage pipeline stream join (Stashed-MJ) that uses a best-effort buffering technique (stash) to maintain intermediate results. In a case that an overwrite is detected in the stash, our design automatically resorts to recomputing intermediate results. Our experimental results demonstrate a linear throughput scaling with respect to the number of execution units in hardware." ++++ diff --git a/content/publications/a-serverless-publish-subscribe-system.md b/content/publications/a-serverless-publish-subscribe-system.md new file mode 100644 index 0000000..5c2e47e --- /dev/null +++ b/content/publications/a-serverless-publish-subscribe-system.md @@ -0,0 +1,10 @@ ++++ +title = "A Serverless Publish/Subscribe System" +year = 2022 +authors = ["Pezhman Nasirifard", "Hans-Arno Jacobsen"] +venue = "arXiv" +publication_type = "ArXiv Preprint" +research = ["data-management"] +external_url = "https://arxiv.org/abs/2210.07897" +abstract = "Operating a scalable and reliable server application, such as publish/subscribe (pub/sub) systems, requires tremendous development efforts and resources. The emerging serverless paradigm simplifies the development and deployment of highly available applications by delegating most operational concerns to the cloud providers. The serverless paradigm describes a programming model where the developers break the application downs into smaller microservices that run on the cloud in response to events. This paper proposes designing a serverless pub/sub system based on the IBM Bluemix cloud platform. Our pub/sub system performs topic-based, content-based, and function-based matchings. The function-based matching is a novel matching approach where the subscribers can define a highly customizable subscription function that the broker applies to the publications in the cloud. The evaluations of the designed system verify the practicality of the designed system. However, the vendor-specific constraints of the IBM Bluemix resources are a bottleneck to the scalability of the broker." ++++ diff --git a/content/publications/a-system-for-semantic-data-fusion-in-sensor-networks.md b/content/publications/a-system-for-semantic-data-fusion-in-sensor-networks.md new file mode 100644 index 0000000..e127080 --- /dev/null +++ b/content/publications/a-system-for-semantic-data-fusion-in-sensor-networks.md @@ -0,0 +1,10 @@ ++++ +title = "A system for semantic data fusion in sensor networks" +year = 2007 +authors = ["Alex Wun", "Milenko Petrovic", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 2007 inaugural international conference on Distributed event-based systems" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/1266894.1266907" +abstract = "Emerging sensor network technologies are expected to substantially augment applications such as environmental monitoring, health-care, and home/commercial automation. However, much of the existing work focuses mainly on collecting and using sensor level data from isolated sensor networks directly, which still burdens applications with the task of interpreting the context and meaning of sensor data. In order to infer high-level phenomena, sensor data needs to be filtered, aggregated, correlated, and translated from many heterogeneous and dispersed sensor networks. In this paper, we present a novel system for decoupling the process of semantic data fusion from application logic based on semantic Content-based Publish/Subscribe techniques. Our main contribution is an integrated system that allows efficient semantic event detection to occur both within and across sensor networks by translating events using ontologies." ++++ diff --git a/content/publications/a-taxonomy-for-denial-of-service-attacks-in-content-based-publish-subscribe-systems.md b/content/publications/a-taxonomy-for-denial-of-service-attacks-in-content-based-publish-subscribe-systems.md new file mode 100644 index 0000000..515cab6 --- /dev/null +++ b/content/publications/a-taxonomy-for-denial-of-service-attacks-in-content-based-publish-subscribe-systems.md @@ -0,0 +1,10 @@ ++++ +title = "A taxonomy for denial of service attacks in content-based publish/subscribe systems" +year = 2007 +authors = ["Alex Wun", "Alex King Yeung Cheung", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 2007 inaugural international conference on Distributed event-based systems" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/1266894.1266917" +abstract = "Denial of Service (DoS) attacks continue to affect the availability of critical systems on the Internet. The existing DoS problem is enough to merit significant research dedicated to analyzing and classifying DoS attacks in the Internet context. However, no such research exists for DoS attacks in the domain of Content-based Publish/Subscribe (CPS) systems despite CPS being at the forefront of business process execution, application integration, and event processing applications. This can be attributed to the lack of structure and understanding of key issues in the area of DoS in CPS systems. In this paper, we propose to address these problems by presenting a taxonomy for classifying DoS characteristics and concerns new to CPS systems. Our taxonomy is motivated by a number of experimental results that were obtained using our CPS middleware implementation and that highlight fundamental DoS concerns in this domain. Finally, we discuss some example DoS attacks in detail with respect to our taxonomy and experimental results. We find that localization, message content complexity, and filter statefulness are the key CPS characteristics to consider when designing DoS resilient CPS systems." ++++ diff --git a/content/publications/a-topss-a-publish-subscribe-system-supporting-approximate-matching.md b/content/publications/a-topss-a-publish-subscribe-system-supporting-approximate-matching.md new file mode 100644 index 0000000..9de485b --- /dev/null +++ b/content/publications/a-topss-a-publish-subscribe-system-supporting-approximate-matching.md @@ -0,0 +1,9 @@ ++++ +title = "A-TOPSS - A Publish/Subscribe System Supporting Approximate Matching" +year = 2002 +authors = ["Haifeng Liu", "Hans-Arno Jacobsen"] +venue = "VLDB '02: Proceedings of the 28th International Conference on Very Large Databases" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1016/b978-155860869-6/50120-7" ++++ diff --git a/content/publications/a-topss-a-publish-subscribe-system-supporting-imperfect-information-processing.md b/content/publications/a-topss-a-publish-subscribe-system-supporting-imperfect-information-processing.md new file mode 100644 index 0000000..24f2f42 --- /dev/null +++ b/content/publications/a-topss-a-publish-subscribe-system-supporting-imperfect-information-processing.md @@ -0,0 +1,9 @@ ++++ +title = "A-ToPSS: A Publish/Subscribe System Supporting Imperfect Information Processing" +year = 2004 +authors = ["Haifeng Liu", "Hans-Arno Jacobsen"] +venue = "Proceedings 2004 VLDB Conference" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1016/b978-012088469-8.50127-3" ++++ diff --git a/content/publications/a-unified-approach-to-routing-covering-and-merging-in-publish-subscribe-systems-based-on-modified-binary-decision-diagrams.md b/content/publications/a-unified-approach-to-routing-covering-and-merging-in-publish-subscribe-systems-based-on-modified-binary-decision-diagrams.md new file mode 100644 index 0000000..5339a46 --- /dev/null +++ b/content/publications/a-unified-approach-to-routing-covering-and-merging-in-publish-subscribe-systems-based-on-modified-binary-decision-diagrams.md @@ -0,0 +1,10 @@ ++++ +title = "A Unified Approach to Routing, Covering and Merging in Publish/Subscribe Systems Based on Modified Binary Decision Diagrams" +year = 2005 +authors = ["Guoli Li", "Shuang Hou", "Hans-Arno Jacobsen"] +venue = "25th IEEE International Conference on Distributed Computing Systems (ICDCS'05)" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icdcs.2005.8" +abstract = "The challenge faced by content-based publish/subscribe systems is the ability to handle a vast amount of dynamic information with limited system resources. In current p/s systems, each subscription is processed in isolation. Neither relationships among individual subscriptions are exploited, nor historic information about subscriptions and publications is taken into account. We believe that this neglect limits overall system efficiency. In this paper, we represent subscriptions using modified binary decision diagrams (MBDs), and design an index data structure to maintain distinct predicates and manage associated Boolean variables. Our MBD-based approach can address, in a unified way, publication routing and subscription/advertisement covering and merging. We propose a novel covering algorithm based on MBDs. The algorithm can take historic information about subscription and publication populations into account and exploits relations between subscriptions. We explore merging, especially imperfect merging, and discuss an advertisement-based optimization applicable to subscription merging" ++++ diff --git a/content/publications/a-user-tunable-machine-learning-framework-for-step-wise-synthesis-planning.md b/content/publications/a-user-tunable-machine-learning-framework-for-step-wise-synthesis-planning.md new file mode 100644 index 0000000..463bcfa --- /dev/null +++ b/content/publications/a-user-tunable-machine-learning-framework-for-step-wise-synthesis-planning.md @@ -0,0 +1,11 @@ ++++ +title = "A user-tunable machine learning framework for step-wise synthesis planning" +year = 2026 +authors = ["Shivesh Prakash", "Nandan Patel", "Hans-Arno Jacobsen", "Viki Kumar Prasad"] +venue = "Digital Discovery" +publication_type = "Journal Article" +research = ["distributed-machine-learning"] +external_url = "https://doi.org/10.1039/d5dd00562k" +abstract = "We introduce MHNpath, a machine learning-driven retrosynthetic tool designed for computer-aided synthesis planning. Leveraging modern Hopfield networks and novel comparative metrics, MHNpath efficiently prioritizes reaction templates, improving the scalability and accuracy of retrosynthetic predictions. The tool incorporates a tunable scoring system that allows users to prioritize pathways based on cost, reaction temperature, and toxicity, thereby facilitating the design of greener and cost-effective reaction routes. We demonstrate its effectiveness through case studies involving complex molecules from ChemByDesign, showcasing its ability to predict novel synthetic and enzymatic pathways. Furthermore, we benchmark MHNpath against existing frameworks using the PaRoutes dataset, achieving a solution rate of 85.4% and replicating 69.2% of experimentally validated “gold-standard” pathways. Our case studies show that the tool can generate shorter, cheaper, moderate-temperature routes employing green solvents, as exemplified by molecules such as dronabinol, arformoterol, and lupinine." +abstract_license_url = "https://creativecommons.org/licenses/by/3.0/" ++++ diff --git a/content/publications/adaptive-content-based-routing-in-general-overlay-topologies.md b/content/publications/adaptive-content-based-routing-in-general-overlay-topologies.md index 03d9cff..676b21c 100644 --- a/content/publications/adaptive-content-based-routing-in-general-overlay-topologies.md +++ b/content/publications/adaptive-content-based-routing-in-general-overlay-topologies.md @@ -3,19 +3,11 @@ title = "Adaptive Content-Based Routing in General Overlay Topologies" slug = "adaptive-content-based-routing-in-general-overlay-topologies" year = 2008 authors = ["Guoli Li", "Vinod Muthusamy", "Hans-Arno Jacobsen"] -venue = "ACM/IFIP/USENIX International Middleware Conference" +venue = "Lecture Notes in Computer Science; Middleware 2008" publication_type = "Conference Paper" research = ["data-management"] tags = ["publish-subscribe", "content-based-routing", "overlay-networks"] -summary = "Develops content-based publish/subscribe algorithms that support general (cyclic) overlay topologies, enabling adaptive routing and composite event detection, implemented in the PADRES system." -external_url = "https://link.springer.com/chapter/10.1007/978-3-540-89856-6_1" +external_url = "https://doi.org/10.1007/978-3-540-89856-6_1" related_datasets = ["cyclic-overlay-workload"] +abstract = "This paper develops content-based publish/subscribe algorithms to support general overlay topologies, as opposed to traditional acyclic or tree-based topologies. Among other benefits, message routes can adapt to dynamic conditions by choosing among alternate routing paths, and composite events can be detected at optimal points in the network. The algorithms are implemented in the PADRES publish/subscribe system and evaluated in a controlled local environment and a wide-area PlanetLab deployment. Atomic subscription notification delivery time improves by 20 % in a well connected network, and composite subscriptions can be processed with 80 % less network traffic and notifications delivered with about half the end to end delay." +++ - -Traditional content-based publish/subscribe systems assume acyclic or tree-based -overlay topologies. This paper develops algorithms for general overlay topologies, -allowing publication routes to adapt to dynamic conditions and composite events to -be detected at optimal points in the network. The implementation in PADRES is -evaluated in both a controlled local environment and a wide-area PlanetLab -deployment, with the companion workload package capturing subscription-generation -patterns for stock-style event streams. diff --git a/content/publications/adaptive-location-constraint-processing.md b/content/publications/adaptive-location-constraint-processing.md new file mode 100644 index 0000000..8069133 --- /dev/null +++ b/content/publications/adaptive-location-constraint-processing.md @@ -0,0 +1,10 @@ ++++ +title = "Adaptive location constraint processing" +year = 2007 +authors = ["Zhengdao Xu", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 2007 ACM SIGMOD international conference on Management of data" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/1247480.1247545" +abstract = "An important problem for many location-based applications is the continuous evaluation of proximity relations among moving objects. These relations express whether a given set of objects is in a spatial constellation or in a spatial constellation relative to a given point of demarcation in the environment. We represent proximity relations as location constraints, which resemble standing queries over continuously changing location position information. The challenge lies in the continuous processing of large numbers of location constraints as the location of objects and the constraint load change. In this paper, we propose an adaptive location constraint indexing approach which adapts as the constraint load and movement pattern of the objects change. The approach takes correlations between constraints into account to further reduce processing time. We also introduce a new location update policy that detects constraint matches with fewer location update requests. Our approach stabilizes system performance, avoids oscillation, reduces constraint matching time by 70% for in-memory processing, and reduces secondary storage accesses by 80% for I/O-incurring environments." ++++ diff --git a/content/publications/adaptive-middleware-for-real-time-prescriptive-analytics-in-large-scale-power-systems.md b/content/publications/adaptive-middleware-for-real-time-prescriptive-analytics-in-large-scale-power-systems.md new file mode 100644 index 0000000..ccf0592 --- /dev/null +++ b/content/publications/adaptive-middleware-for-real-time-prescriptive-analytics-in-large-scale-power-systems.md @@ -0,0 +1,10 @@ ++++ +title = "Adaptive middleware for real-time prescriptive analytics in large scale power systems" +year = 2013 +authors = ["Sebnem Rusitschka", "Christoph Doblander", "Christoph Goebel", "Hans-Arno Jacobsen"] +venue = "Proceedings of the Industrial Track of the 13th ACM/IFIP/USENIX International Middleware Conference" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/2541596.2541601" +abstract = "The increased digitalization of power systems poses both opportunities and challenges for system operators. GPS time-synchronized high-resolution data streams emanating from measurement devices distributed over a wide area enable the detection of disturbances and the real-time monitoring of consequences as they are evolving, such as undamped oscillations. Processing these data streams is not possible with state-of-the-art SCADA systems that poll data asynchronously at much lower time intervals. Moreover, real-time analysis on fresh streaming data at the enterprise level is an unresolved challenge. In this paper we propose an adaptive middleware concept that can make better use of available data processing resources by enabling distributed computation both on the enterprise and on the field level. We apply the concept of linked data to provide a map for moving the computation to the data it requires for analysis. If based on the IEC 61850 standard semantic data model, the linked data concept additionally yields location and domain awareness that can be leveraged for real-time prescriptive analytics in the field. Another advantage of the proposed adaptive middleware is the abstraction of computational resources: Analytical programs can be written once and then be used to process historical data residing on servers on the enterprise level as well on the distributed devices that originated the data to enable fast analysis of events as they are unfolding." ++++ diff --git a/content/publications/adaptive-parallel-compressed-event-matching.md b/content/publications/adaptive-parallel-compressed-event-matching.md new file mode 100644 index 0000000..fa569dd --- /dev/null +++ b/content/publications/adaptive-parallel-compressed-event-matching.md @@ -0,0 +1,10 @@ ++++ +title = "Adaptive parallel compressed event matching" +year = 2014 +authors = ["Mohammad Sadoghi", "Hans-Arno Jacobsen"] +venue = "2014 IEEE 30th International Conference on Data Engineering" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icde.2014.6816665" +abstract = "The efficient processing of large collections of patterns expressed as Boolean expressions over event streams plays a central role in major data intensive applications ranging from user-centric processing and personalization to real-time data analysis. On the one hand, emerging user-centric applications, including computational advertising and selective information dissemination, demand determining and presenting to an end-user the relevant content as it is published. On the other hand, applications in real-time data analysis, including push-based multi-query optimization, computational finance and intrusion detection, demand meeting stringent subsecond processing requirements and providing high-frequency event processing. We achieve these event processing requirements by exploiting the shift towards multi-core architectures by proposing novel adaptive parallel compressed event matching algorithm (A-PCM) and online event stream re-ordering technique (OSR) that unleash an unprecedented degree of parallelism amenable for highly parallel event processing. In our comprehensive evaluation, we demonstrate the efficiency of our proposed techniques. We show that the adaptive parallel compressed event matching algorithm can sustain an event rate of up to 233,863 events/second while state-of-the-art sequential event matching algorithms sustains only 36 events/second when processing up to five million Boolean expressions." ++++ diff --git a/content/publications/adversarial-robustness-in-distributed-quantum-machine-learning.md b/content/publications/adversarial-robustness-in-distributed-quantum-machine-learning.md index 2593086..e54a268 100644 --- a/content/publications/adversarial-robustness-in-distributed-quantum-machine-learning.md +++ b/content/publications/adversarial-robustness-in-distributed-quantum-machine-learning.md @@ -1,12 +1,11 @@ +++ title = "Adversarial Robustness in Distributed Quantum Machine Learning" -year = 2025 +year = 2026 authors = ["Pouya Kananian", "Hans-Arno Jacobsen"] -venue = "arXiv" -publication_type = "ArXiv Preprint" +venue = "Quantum Science and Technology; Quantum Robustness in Artificial Intelligence" +publication_type = "Book Chapter" research = ["quantum-computing-systems"] tags = ["quantum-systems"] -external_url = "https://arxiv.org/abs/2508.11848" -source_url = "https://arxiv.org/abs/2508.11848" +external_url = "https://doi.org/10.1007/978-3-032-11153-1_11" url = "/publications/adversarial-robustness-in-distributed-quantum-machine-learning/" +++ diff --git a/content/publications/adversarial-robustness-of-partitioned-quantum-classifiers.md b/content/publications/adversarial-robustness-of-partitioned-quantum-classifiers.md index cc9f666..04d66ae 100644 --- a/content/publications/adversarial-robustness-of-partitioned-quantum-classifiers.md +++ b/content/publications/adversarial-robustness-of-partitioned-quantum-classifiers.md @@ -7,5 +7,4 @@ publication_type = "ArXiv Preprint" research = ["quantum-computing-systems"] tags = ["quantum-systems"] external_url = "https://arxiv.org/abs/2502.20403" -source_url = "https://arxiv.org/abs/2502.20403" +++ diff --git a/content/publications/aggregator-controlled-ev-charging-in-pay-as-bid-reserve-markets-with-strict-delivery-constraints.md b/content/publications/aggregator-controlled-ev-charging-in-pay-as-bid-reserve-markets-with-strict-delivery-constraints.md new file mode 100644 index 0000000..bd3b2dd --- /dev/null +++ b/content/publications/aggregator-controlled-ev-charging-in-pay-as-bid-reserve-markets-with-strict-delivery-constraints.md @@ -0,0 +1,10 @@ ++++ +title = "Aggregator-Controlled EV Charging in Pay-as-Bid Reserve Markets With Strict Delivery Constraints" +year = 2016 +authors = ["Christoph Goebel", "Hans-Arno Jacobsen"] +venue = "IEEE Transactions on Power Systems" +publication_type = "Journal Article" +research = [] +external_url = "https://doi.org/10.1109/tpwrs.2016.2518648" +abstract = "Electric vehicles (EVs), similar to other types of flexible loads, can be controlled to provide additional flexibility to power system operators, which is needed to transition from primarily fossil fuel based electricity generation to more renewable generation. Existing electricity markets already provide economic incentives to offer flexibility. So-called aggregators in charge of controlling EV charging could take advantage of these incentives. However, random driving behavior and area-specific market rules complicate the market participation of EV aggregators in energy and reserve markets. The goal of this paper is to investigate the design and performance of a system that would enable EV aggregators to participate in wholesale electricity markets. We consider a certain type of market environment, which includes an intraday energy market and pay-as-bid reserve markets that require exact reserve delivery and have long operating intervals. We therefore propose a novel approach for concurrent market participation that can deal with these constraints, even when faced with highly uncertain reserve activation and EV behavior." ++++ diff --git a/content/publications/algorithms-based-on-divide-and-conquer-for-topic-based-publish-subscribe-overlay-design.md b/content/publications/algorithms-based-on-divide-and-conquer-for-topic-based-publish-subscribe-overlay-design.md new file mode 100644 index 0000000..573a7ec --- /dev/null +++ b/content/publications/algorithms-based-on-divide-and-conquer-for-topic-based-publish-subscribe-overlay-design.md @@ -0,0 +1,10 @@ ++++ +title = "Algorithms Based on Divide and Conquer for Topic-Based Publish/Subscribe Overlay Design" +year = 2016 +authors = ["Chen Chen", "Hans-Arno Jacobsen", "Roman Vitenberg"] +venue = "IEEE/ACM Transactions on Networking" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.1109/tnet.2014.2369346" +abstract = "Overlay design for topic-based publish/subscribe (pub/sub) systems is of primary importance because the overlay forms the basis for the system and directly impacts its performance. This paper focuses on the MinAvg-TCO problem: Use the minimum number of edges to construct a topic-connected overlay (TCO) such that all nodes that are interested in the same topic are organized in a directly connected dissemination suboverlay. Existing algorithms for MinAvg-TCO suffer from three key drawbacks: 1) prohibitively high runtime cost; 2) reliance on global knowledge and centralized operation; and 3) nonincremental operation by reconstructing the TCO from scratch. From a practical point of view, these are all severe limitations. To address these concerns, we develop algorithms that dynamically join multiple TCOs. Inspired by the divide-and-conquer character of this idea, we derive a number of algorithms for the original MinAvg-TCO problem that accommodate a variety of practical pub/sub workloads. Both theoretical analysis and experimental evaluations demonstrate that our divide-and-conquer algorithms seek a balance between time efficiency and the number of edges required: Our algorithms cost a fraction (up to 1.67%) of the runtime cost of their greedy alternatives, which come at the expense of an empirically insignificant increase in the average node degree. Furthermore, in order to reduce the probability of poor partitioning at the divide phase, we develop a bulk-lightweight partitioning scheme on top of random partitioning. This more refined partitioning imposes a marginally higher runtime cost, but leads to improvements in the output TCOs, including average node degrees and topic diameters." ++++ diff --git a/content/publications/alternating-direction-method-of-multipliers-for-decentralized-electric-vehicle-charging-control.md b/content/publications/alternating-direction-method-of-multipliers-for-decentralized-electric-vehicle-charging-control.md new file mode 100644 index 0000000..8f2efe3 --- /dev/null +++ b/content/publications/alternating-direction-method-of-multipliers-for-decentralized-electric-vehicle-charging-control.md @@ -0,0 +1,10 @@ ++++ +title = "Alternating Direction Method of Multipliers for decentralized electric vehicle charging control" +year = 2013 +authors = ["José Rivera", "Philipp Wolfrum", "Sandra Hirche", "Christoph Goebel", "Hans-Arno Jacobsen"] +venue = "52nd IEEE Conference on Decision and Control" +publication_type = "Conference Paper" +research = [] +external_url = "https://doi.org/10.1109/cdc.2013.6760992" +abstract = "The integration of Electric Vehicles (EVs) into the power grid is a challenging task. From the control perspective, one of the main challenges is the definition of a comprehensive control structure that is scalable to large EV numbers. This paper makes two key contributions: (i) It defines the EV ADMM framework for decentralized EV charging control. (ii) It evaluates EV ADMM using actual data and various EV fleet control problems. EV ADMM is a decentralized optimization algorithm based on the Alternating Direction Method of Multipliers (ADMM). It separates the centralized optimal fleet charging problem into individual optimization problems for the EVs plus one aggregator problem that optimizes fleet goals. Since the individual problems are coupled, they are solved consistently by passing incentive signals between them. The framework can be parameterized to trade-off the importance of fleet goals versus individual EV goals, such that aspects like battery lifetime can be considered. We show how EV ADMM can be applied to control an EV fleet to achieve goals such as demand valley filling and minimal-cost charging. Due to its flexibility and scalability, EV ADMM offers a practicable solution for optimal EV fleet control." ++++ diff --git a/content/publications/an-end-to-end-performance-comparison-of-seven-permissioned-blockchain-systems.md b/content/publications/an-end-to-end-performance-comparison-of-seven-permissioned-blockchain-systems.md new file mode 100644 index 0000000..6f23392 --- /dev/null +++ b/content/publications/an-end-to-end-performance-comparison-of-seven-permissioned-blockchain-systems.md @@ -0,0 +1,10 @@ ++++ +title = "An End-to-End Performance Comparison of Seven Permissioned Blockchain Systems" +year = 2023 +authors = ["Frank Christian Geyer", "Hans-Arno Jacobsen", "Ruben Mayer", "Peter Mandl"] +venue = "Proceedings of the 24th International Middleware Conference on ZZZ" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/3590140.3629106" +abstract = "The emergence of numerous blockchain solutions, offering innovative approaches to optimise performance, scalability, privacy, and governance, complicates performance analysis. Reasons for the difficulty of benchmarking blockchains include, for example, the high number of system parameters to configure and the effort to deploy a blockchain network. In addition, performance data, which mostly comes from system vendors, is often opaque. We provide a reproducible evaluation of the performance of seven permissioned blockchain systems across different parameter settings. We employ an end-to-end approach, where the clients sending the transactions are fully involved in the data collection approach. Our results underscore the unique characteristics and limitations of the systems we examined. Due to the insights given, our work forms the basis for continued research to optimise the performance of blockchain systems." ++++ diff --git a/content/publications/an-experimental-comparison-of-partitioning-strategies-for-distributed-graph-neural-network-training.md b/content/publications/an-experimental-comparison-of-partitioning-strategies-for-distributed-graph-neural-network-training.md index 563a8c4..66da8d8 100644 --- a/content/publications/an-experimental-comparison-of-partitioning-strategies-for-distributed-graph-neural-network-training.md +++ b/content/publications/an-experimental-comparison-of-partitioning-strategies-for-distributed-graph-neural-network-training.md @@ -2,10 +2,10 @@ title = "An Experimental Comparison of Partitioning Strategies for Distributed Graph Neural Network Training" year = 2025 authors = ["Nikolai Merkel", "Daniel Stoll", "Ruben Mayer", "Hans-Arno Jacobsen"] -venue = "Proceedings of the 28th International Conference on Extending Database Technology" +venue = "EDBT" publication_type = "Conference Paper" research = ["distributed-machine-learning"] tags = ["graph-systems"] -external_url = "https://openproceedings.org/2025/conf/edbt/paper-22.pdf" -source_url = "https://msrg.utoronto.ca/publications/" +external_url = "https://doi.org/10.48786/edbt.2025.14" +abstract = "Recently, graph neural networks (GNNs) have gained much attention as a growing area of deep learning capable of learning on graph-structured data. However, the computational and memory requirements for training GNNs on large-scale graphs make it necessary to distribute the training. A prerequisite for distributed GNN training is to partition the input graph into smaller parts that are distributed among multiple machines of a compute cluster. Although graph partitioning has been studied with regard to graph analytics and graph databases, its effect on GNN training performance is largely unexplored. As a consequence, it is unclear whether investing computational efforts into high-quality graph partitioning would pay off in GNN training scenarios. In this paper, we study the effectiveness of graph partitioning for distributed GNN training. Our study aims to understand how different factors such as GNN parameters, mini-batch size, graph type, features size, and scale-out factor influence the effectiveness of graph partitioning. We conduct experiments with two different GNN systems using vertex and edge partitioning. We found that high-quality graph partitioning is a very effective optimization to speed up GNN training and to reduce memory consumption. Furthermore, our results show that invested partitioning time can quickly be amortized by reduced GNN training time, making it a relevant optimization for most GNN scenarios. Compared to research on distributed graph processing, our study reveals that graph partitioning plays an even more significant role in distributed GNN training, which motivates further research on the graph partitioning problem." +++ diff --git a/content/publications/analysis-and-optimization-for-boolean-expression-indexing.md b/content/publications/analysis-and-optimization-for-boolean-expression-indexing.md new file mode 100644 index 0000000..3761197 --- /dev/null +++ b/content/publications/analysis-and-optimization-for-boolean-expression-indexing.md @@ -0,0 +1,10 @@ ++++ +title = "Analysis and optimization for boolean expression indexing" +year = 2013 +authors = ["Mohammad Sadoghi", "Hans-Arno Jacobsen"] +venue = "ACM Transactions on Database Systems" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.1145/2487259.2487260" +abstract = "BE-Tree is a novel dynamic data structure designed to efficiently index Boolean expressions over a high-dimensional discrete space. BE Tree-copes with both high-dimensionality and expressiveness of Boolean expressions by introducing an effective two-phase space-cutting technique that specifically utilizes the discrete and finite domain properties of the space. Furthermore, BE-Tree employs self-adjustment policies to dynamically adapt the tree as the workload changes. Moreover, in BE-Tree, we develop two novel cache-conscious predicate evaluation techniques, namely, lazy and bitmap evaluations, that also exploit the underlying discrete and finite space to substantially reduce BE-Tree's matching time by up to 75% BE-Tree is a general index structure for matching Boolean expression which has a wide range of applications including (complex) event processing, publish/subscribe matching, emerging applications in cospaces, profile matching for targeted web advertising, and approximate string matching. Finally, the superiority of BE-Tree is proven through a comprehensive evaluation with state-of-the-art index structures designed for matching Boolean expressions." ++++ diff --git a/content/publications/analysis-of-tpc-ds-the-first-standard-benchmark-for-sql-based-big-data-systems.md b/content/publications/analysis-of-tpc-ds-the-first-standard-benchmark-for-sql-based-big-data-systems.md new file mode 100644 index 0000000..0a011d8 --- /dev/null +++ b/content/publications/analysis-of-tpc-ds-the-first-standard-benchmark-for-sql-based-big-data-systems.md @@ -0,0 +1,10 @@ ++++ +title = "Analysis of TPC-DS: the first standard benchmark for SQL-based big data systems" +year = 2017 +authors = ["Meikel Poess", "Tilmann Rabl", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 2017 Symposium on Cloud Computing" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/3127479.3128603" +abstract = "The advent of Web 2.0 companies, such as Facebook, Google, and Amazon with their insatiable appetite for vast amounts of structured, semi-structured, and unstructured data, triggered the development of Hadoop and related tools, e.g., YARN, MapReduce, and Pig, as well as NoSQL databases. These tools form an open source software stack to support the processing of large and diverse data sets on clustered systems to perform decision support tasks. Recently, SQL is resurrecting in many of these solutions, e.g., Hive, Stinger, Impala, Shark, and Presto. At the same time, RDBMS vendors are adding Hadoop support into their SQL engines, e.g., IBM's Big SQL, Actian's Vortex, Oracle's Big Data SQL, and SAP's HANA. Because there was no industry standard benchmark that could measure the performance of SQL-based big data solutions, marketing claims were mostly based on \"cherry picked\" subsets of the TPC-DS benchmark to suit individual companies strengths, while blending out their weaknesses. In this paper, we present and analyze our work on modifying TPC-DS to fill the void for an industry standard benchmark that is able to measure the performance of SQL-based big data solutions. The new benchmark was ratified by the TPC in early 2016. To show the significance of the new benchmark, we analyze performance data obtained on four different systems running big data, traditional RDBMS, and columnar in-memory architectures." ++++ diff --git a/content/publications/analysis-of-tpcx-iot-the-first-industry-standard-benchmark-for-iot-gateway-systems.md b/content/publications/analysis-of-tpcx-iot-the-first-industry-standard-benchmark-for-iot-gateway-systems.md new file mode 100644 index 0000000..5c81b29 --- /dev/null +++ b/content/publications/analysis-of-tpcx-iot-the-first-industry-standard-benchmark-for-iot-gateway-systems.md @@ -0,0 +1,10 @@ ++++ +title = "Analysis of TPCx-IoT: The First Industry Standard Benchmark for IoT Gateway Systems" +year = 2018 +authors = ["Meikel Poess", "Raghunath Nambiar", "Karthik Kulkarni", "Chinmayi Narasimhadevara", "Tilmann Rabl", "Hans-Arno Jacobsen"] +venue = "2018 IEEE 34th International Conference on Data Engineering (ICDE)" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icde.2018.00170" +abstract = "By 2020 it is estimated that 20 billion devices will be connected to the Internet. While the initial hype around this Internet of Things (IoT) stems from consumer use cases, the number of devices and data from enterprise use cases is significant in terms of market share. With companies being challenged to choose the right digital infrastructure from different providers, there is an pressing need to objectively measure the hardware, operating system, data storage, and data management systems that can ingest, persist, and process the massive amounts of data arriving from sensors (edge devices). The Transaction Processing Performance Council (TPC) recently released the first industry standard benchmark for measuring the performance of gateway systems, TPCx-IoT. In this paper, we provide a detailed description of TPCx-IoT, mention design decisions behind key elements of this benchmark, and experimentally analyze how TPCx-IoT measures the performance of IoT gateway systems." ++++ diff --git a/content/publications/analyzing-common-electronic-structure-theory-algorithms-for-distributed-quantum-computing.md b/content/publications/analyzing-common-electronic-structure-theory-algorithms-for-distributed-quantum-computing.md index 3087aac..8fd2144 100644 --- a/content/publications/analyzing-common-electronic-structure-theory-algorithms-for-distributed-quantum-computing.md +++ b/content/publications/analyzing-common-electronic-structure-theory-algorithms-for-distributed-quantum-computing.md @@ -2,10 +2,10 @@ title = "Analyzing Common Electronic Structure Theory Algorithms for Distributed Quantum Computing" year = 2025 authors = ["Grier M. Jones", "Hans-Arno Jacobsen"] -venue = "arXiv" -publication_type = "ArXiv Preprint" +venue = "2025 IEEE International Conference on Quantum Computing and Engineering (QCE)" +publication_type = "Conference Paper" research = ["quantum-computing-systems"] tags = ["quantum-systems"] -external_url = "https://arxiv.org/abs/2507.01902" -source_url = "https://arxiv.org/abs/2507.01902" +external_url = "https://doi.org/10.1109/qce65121.2025.10351" +abstract = "To move towards the utility era of quantum computing, many corporations have posed distributed quantum computing (DQC) as a framework for scaling the current generation of devices for practical applications. One of these applications is quantum chemistry, also known as electronic structure theory, which has been poised as a “killer application” of quantum computing. To this end, we analyze five electronic structure methods, including the unitary coupled-cluster singles and doubles (UCCSD), unitary pair coupled-cluster doubles (UpCCD), unitary pair coupled-cluster with generalized singles and doubles (UpCCGSD), separable pair approximation plus generalized singles (SPA+GS), and local unitary cluster Jastrow (LUCJ) ansätze. The key benefit of these methods is that they can be found in common packages, such as Tequila and ffsim, that interface easily with the Qiskit Circuit Cutting addon. Herein, we highlight the challenges and aptitude of the aforementioned electronic structure methods for distribution using quasiprobability simulation with local operations (LO). The key findings of our work are that many of these algorithms cannot be efficiently parallelized using LO, and new methods must be developed to apply electronic structure theory within a DQC framework." +++ diff --git a/content/publications/analyzing-geospatial-distribution-in-blockchains.md b/content/publications/analyzing-geospatial-distribution-in-blockchains.md index defdf25..32c09f5 100644 --- a/content/publications/analyzing-geospatial-distribution-in-blockchains.md +++ b/content/publications/analyzing-geospatial-distribution-in-blockchains.md @@ -6,6 +6,6 @@ venue = "2023 IEEE International Conference on Decentralized Applications and In publication_type = "Conference Paper" research = ["data-management"] tags = ["blockchain"] -external_url = "https://ieeexplore.ieee.org/document/10237020/" -source_url = "https://msrg.utoronto.ca/publications/" +external_url = "https://doi.org/10.1109/dapps57946.2023.00022" +abstract = "Blockchains are decentralized; are they genuinely? We analyze blockchain decentralization's often-overlooked but quantifiable dimension: geospatial distribution of transaction processing. Blockchains bring with them the potential for geospatially distributed transaction processing. They enable validators from geospatially distant locations to partake in consensus protocols; we refer to them as minority validators. Based on our observations, in practice, most validators are often geographically concentrated in close proximity. Furthermore, we observed that minority validators tend not to meet the performance requirements, often misidentified as crash failures. Consequently, they are subject to punishment by jailing (removal from the validator set) and/or slashing (penalty in native tokens). Our emulations, under controlled conditions, demonstrate the same results, raising serious concerns about the potential for the geospatial centralization of validators. To address this, we developed a solution that easily integrates with consensus protocols, and we demonstrated its effectiveness." +++ diff --git a/content/publications/analyzing-initialization-strategies-for-the-local-unitary-cluster-jastrow-ansatz-within-the-quantum-centric-supercomputing-framework.md b/content/publications/analyzing-initialization-strategies-for-the-local-unitary-cluster-jastrow-ansatz-within-the-quantum-centric-supercomputing-framework.md new file mode 100644 index 0000000..59ebc02 --- /dev/null +++ b/content/publications/analyzing-initialization-strategies-for-the-local-unitary-cluster-jastrow-ansatz-within-the-quantum-centric-supercomputing-framework.md @@ -0,0 +1,10 @@ ++++ +title = "Analyzing Initialization Strategies for the Local Unitary Cluster Jastrow Ansatz within the Quantum-Centric Supercomputing Framework" +year = 2026 +authors = ["Grier M. Jones", "Maforikan J. Amoussou", "Maximilian O. Leach", "Hans-Arno Jacobsen"] +venue = "arXiv" +publication_type = "ArXiv Preprint" +research = ["quantum-computing-systems"] +external_url = "https://arxiv.org/abs/2606.14933" +abstract = "In this study, we analyze the choice of local unitary cluster Jastrow (LUCJ) ansatz initialization and sensitivity of the sample-based quantum diagonalization (SQD) algorithm within the quantum-centric supercomputing (QCSC) framework. We examine six initialization strategies, including those based on coupled-cluster singles and doubles (CCSD), Møller-Plesset second-order perturbation theory (MP2), data-driven coupled-cluster (DDCC), and trivial (zeroes and random) initializations, across twelve molecular systems and three basis sets (STO-3G, cc-pVDZ, and aug-cc-pVDZ). We find that while the mean absolute percentage errors (MAPEs) between the alternative and CCSD-initialized t2-amplitudes span many orders of magnitude, the resulting SQD energies are largely insensitive to this variation. In particular, most initializations recover energies within chemical accuracy (+/-1.6 mEh) of the CCSD reference, with convergence improving as the basis set size increases. Notably, random initialization achieves performance competitive with CCSD across all basis sets, while zeroes initialization, despite having smaller deviations from CCSD, yields the worst energy agreement. Our results highlight that the proximity to the CCSD initialization is not a reliable predictor of the quality of electronic energies. These findings establish that configuration recovery within SQD, rather than circuit initialization, is the dominant factor governing energy accuracy, and suggest that computationally cheaper initialization strategies are viable alternatives to CCSD for QCSC workflows" ++++ diff --git a/content/publications/appliance-classification-across-multiple-high-frequency-energy-datasets.md b/content/publications/appliance-classification-across-multiple-high-frequency-energy-datasets.md new file mode 100644 index 0000000..b4ed70d --- /dev/null +++ b/content/publications/appliance-classification-across-multiple-high-frequency-energy-datasets.md @@ -0,0 +1,10 @@ ++++ +title = "Appliance classification across multiple high frequency energy datasets" +year = 2017 +authors = ["Matthias Kahl", "Thomas Kriechbaumer", "Anwar Ul Haq", "Hans-Arno Jacobsen"] +venue = "2017 IEEE International Conference on Smart Grid Communications (SmartGridComm)" +publication_type = "Conference Paper" +research = [] +external_url = "https://doi.org/10.1109/smartgridcomm.2017.8340664" +abstract = "Non-intrusive load monitoring (NILM) provides several techniques for demand information retrieval to support consumers saving energy usage. Research in NILM often focuses on closed environments, such as single datasets or single households. Disaggregation results are typically not suitable to represent the classification performance under real circumstances due to its data homogeneity of a single dataset. We apply a classification system across four commonly available high frequency energy datasets. The experiments include classification tasks with four different classifiers on 36 spectral and temporal features to perform a cross-, mixed-, and intra-dataset validation. The outcome of this work is a reliable benchmark for appliance recognition in the high frequency domain and its efficiency in smart meters for different use cases and appliance features." ++++ diff --git a/content/publications/appliance-event-detection-a-multivariate-supervised-classification-approach.md b/content/publications/appliance-event-detection-a-multivariate-supervised-classification-approach.md new file mode 100644 index 0000000..9511d60 --- /dev/null +++ b/content/publications/appliance-event-detection-a-multivariate-supervised-classification-approach.md @@ -0,0 +1,10 @@ ++++ +title = "Appliance Event Detection - A Multivariate, Supervised Classification Approach" +year = 2019 +authors = ["Matthias Kahl", "Thomas Kriechbaumer", "Daniel Jorde", "Anwar Ul Haq", "Hans-Arno Jacobsen"] +venue = "Proceedings of the Tenth ACM International Conference on Future Energy Systems" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/3307772.3330155" +abstract = "Appliance event detection is an elementary step in the NILM pipeline. Unfortunately, several types of appliances (e.g., switching mode power supply (SMPS) or multi-state) are known to challenge state-of-the-art event detection systems due to their noisy consumption profiles. By stepping away from distinct event definitions, we learn from a consumer-configured event model to differentiate between relevant and irrelevant event transients. We introduce a boosting oriented adaptive training, that uses false positives from the initial training area to reduce the number of false positives on the test area substantially. The results show a false positive decrease by more than a factor of eight on a dataset that has a strong focus on SMPS-driven appliances. To obtain a stable event detection system, we applied many experiments on different parameters to measure its performance on two publicly available energy datasets." ++++ diff --git a/content/publications/applications-of-noisy-quantum-computing-and-quantum-error-mitigation-to-adamantaneland-a-benchmarking-study-for-quantum-chemistry.md b/content/publications/applications-of-noisy-quantum-computing-and-quantum-error-mitigation-to-adamantaneland-a-benchmarking-study-for-quantum-chemistry.md new file mode 100644 index 0000000..c362c3e --- /dev/null +++ b/content/publications/applications-of-noisy-quantum-computing-and-quantum-error-mitigation-to-adamantaneland-a-benchmarking-study-for-quantum-chemistry.md @@ -0,0 +1,10 @@ ++++ +title = "Applications of noisy quantum computing and quantum error mitigation to “adamantaneland”: a benchmarking study for quantum chemistry" +year = 2024 +authors = ["Viki Kumar Prasad", "Freeman Cheng", "Ulrich Fekl", "Hans-Arno Jacobsen"] +venue = "Physical Chemistry Chemical Physics" +publication_type = "Journal Article" +research = ["quantum-computing-systems"] +external_url = "https://doi.org/10.1039/d3cp03523a" +abstract = ", calculated at a high-level of theory. Using the data set, we compared noiseless VQE simulations to conventionally performed density functional and wavefunction theory-based methods to understand the quality of results. We also investigated the effectiveness of a quantum state tomography-based error mitigation technique in applications of VQE under noise (simulated and real). Our findings reveal that the use of quantum error mitigation is crucial in the NISQ era and advantageous to yield almost noiseless quality results." ++++ diff --git a/content/publications/approximate-matching-in-publish-subscribe.md b/content/publications/approximate-matching-in-publish-subscribe.md new file mode 100644 index 0000000..e8a253e --- /dev/null +++ b/content/publications/approximate-matching-in-publish-subscribe.md @@ -0,0 +1,10 @@ ++++ +title = "Approximate matching in publish/ subscribe" +year = 2003 +authors = ["Haifeng Liu", "Hans-Arno Jacobsen"] +venue = "Proceedings 2003 IEEE International Symposium on Computational Intelligence in Robotics and Automation. Computational Intelligence in Robotics and Automation for the New Millennium (Cat. No.03EX694)" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/cira.2003.1222087" +abstract = "The publish/subscribe paradigm has found wide-spread applications, including selective information dissemination, location-based services, enterprises application integration, and network management. However, all existing publish/subscribe system models cannot capture any kind of uncertainty naturally inherent to many real world scenarios about formulated. To address this shortcoming, this paper proposes a new publish/subscribe system model to process uncertainties in both subscriptions and publications. The system model is evaluated in an implementation of a publish/subscribe system supporting uncertainties in publications and subscriptions through an approximate matching semantic." ++++ diff --git a/content/publications/automatic-generation-of-real-power-transmission-grid-models-from-crowdsourced-data.md b/content/publications/automatic-generation-of-real-power-transmission-grid-models-from-crowdsourced-data.md new file mode 100644 index 0000000..a411ead --- /dev/null +++ b/content/publications/automatic-generation-of-real-power-transmission-grid-models-from-crowdsourced-data.md @@ -0,0 +1,10 @@ ++++ +title = "Automatic Generation of Real Power Transmission Grid Models From Crowdsourced Data" +year = 2019 +authors = ["José Rivera", "Pezhman Nasirifard", "Johannes Leimhofer", "Hans-Arno Jacobsen"] +venue = "IEEE Transactions on Smart Grid" +publication_type = "Journal Article" +research = [] +external_url = "https://doi.org/10.1109/tsg.2018.2882840" +abstract = "Real models of electrical transmission grids are difficult to obtain. The process of generating such models from unstructured and incomplete data is tedious, and the resulting models are rarely updated. This paper proposes a novel approach for automatically extracting power-relevant data from the public and unstructured crowdsourced OpenStreetMap (OSM) and for generating topology and simulation-ready models of real transmission grids based on the relation between different grid elements such as power lines and substations. Our approach uses spatial analysis and minor assumptions to periodically generate transmission grid models based on the latest OSM data for every country on the planet. A comparison of our generated power grid models with official data from 14 countries reveals accuracy levels between 31% and 94%, caused by the varying availability of OSM data for different countries. Since the crowdsourced data is continuously improving, the automated and periodical model generation approach extends the models with new power circuits as the quantity and the quality of the OSM dataset increases. We provide a platform to access our generated models at open-gridmap.org. This paper describes our model generation method, presents our data access platform and evaluates the accuracy of our topological models for selected countries." ++++ diff --git a/content/publications/automating-sla-modeling.md b/content/publications/automating-sla-modeling.md new file mode 100644 index 0000000..347791e --- /dev/null +++ b/content/publications/automating-sla-modeling.md @@ -0,0 +1,10 @@ ++++ +title = "Automating SLA modeling" +year = 2008 +authors = ["Tony Chau", "Vinod Muthusamy", "Hans-Arno Jacobsen", "Elena Litani", "Allen Chan", "Phil Coulthard"] +venue = "Proceedings of the 2008 conference of the center for advanced studies on collaborative research meeting of minds - CASCON '08" +publication_type = "Conference Paper" +research = [] +external_url = "https://doi.org/10.1145/1463788.1463802" +abstract = "Service Level Agreements (SLAs) define the level of service that a service provider must deliver. An SLA is a contract between service provider and consumer, and includes appropriate actions to be taken upon violation of the contractual obligations. However, implementing an SLA using existing IT infrastructure is difficult, requiring a lot of manual effort to translate an SLA into code, model it with the given programming language, and ensure the required monitoring support is available for efficient monitoring and tracking of the SLAs." ++++ diff --git a/content/publications/balancing-power-and-performance-in-hpc-clouds.md b/content/publications/balancing-power-and-performance-in-hpc-clouds.md new file mode 100644 index 0000000..c5bab78 --- /dev/null +++ b/content/publications/balancing-power-and-performance-in-hpc-clouds.md @@ -0,0 +1,10 @@ ++++ +title = "Balancing Power And Performance In HPC Clouds" +year = 2020 +authors = ["Lixia Chen", "Jian Li", "Ruhui Ma", "Haibing Guan", "Hans-Arno Jacobsen"] +venue = "The Computer Journal" +publication_type = "Journal Article" +research = [] +external_url = "https://doi.org/10.1093/comjnl/bxz150" +abstract = "With energy consumption in high-performance computing clouds growing rapidly, energy saving has become an important topic. Virtualization provides opportunities to save energy by enabling one physical machine (PM) to host multiple virtual machines (VMs). Dynamic voltage and frequency scaling (DVFS) is another technology to reduce energy consumption. However, in heterogeneous cloud environments where DVFS may be applied at the chip level or the core level, it is a great challenge to combine these two technologies efficiently. On per-core DVFS servers, cloud managers should carefully determine VM placements to minimize performance interference. On full-chip DVFS servers, cloud managers further face the choice of whether to combine VMs with different characteristics to reduce performance interference or to combine VMs with similar characteristics to take better advantage of DVFS. This paper presents a novel mechanism combining a VM placement algorithm and a frequency scaling method. We formulate this VM placement problem as an integer programming (IP) to find appropriate placement configurations, and we utilize support vector machines to select suitable frequencies. We conduct detailed experiments and simulations, showing that our scheme effectively reduces energy consumption with modest impact on performance. Particularly, the total energy delay product is reduced by up to 60%." ++++ diff --git a/content/publications/be-tree-an-index-structure-for-boolean-expression-matching.md b/content/publications/be-tree-an-index-structure-for-boolean-expression-matching.md index 653f124..8dfe43b 100644 --- a/content/publications/be-tree-an-index-structure-for-boolean-expression-matching.md +++ b/content/publications/be-tree-an-index-structure-for-boolean-expression-matching.md @@ -1,19 +1,13 @@ +++ -title = "Be-tree: An Index Structure to Efficiently Match Boolean Expressions over High-Dimensional Discrete Space" +title = "BE-tree: an index structure to efficiently match boolean expressions over high-dimensional discrete space" slug = "be-tree-an-index-structure-for-boolean-expression-matching" year = 2011 authors = ["Mohammad Sadoghi", "Hans-Arno Jacobsen"] -venue = "Proceedings of the 2011 ACM SIGMOD International Conference on Management of Data" +venue = "Proceedings of the 2011 ACM SIGMOD International Conference on Management of data" publication_type = "Conference Paper" research = ["data-management"] tags = ["boolean-expression-matching", "publish-subscribe", "indexing"] -summary = "Introduces the BE-Tree, a tree-based index for efficiently matching large sets of Boolean expressions over high-dimensional discrete attribute spaces, with the companion BEGen workload generator used to evaluate it." -external_url = "https://dl.acm.org/doi/abs/10.1145/1989323.1989390" +external_url = "https://doi.org/10.1145/1989323.1989390" related_datasets = ["begen"] +abstract = "BE-Tree is a novel dynamic tree data structure designed to efficiently index Boolean expressions over a high-dimensional discrete space. BE-Tree copes with both high-dimensionality and expressiveness of Boolean expressions by introducing a novel two-phase space-cutting technique that specifically utilizes the discrete and finite domain properties of the space. Furthermore, BE-Tree employs self-adjustment policies to dynamically adapt the tree as the workload changes. We conduct a comprehensive evaluation to demonstrate the superiority of BE-Tree in comparison with state-of-the-art index structures designed for matching Boolean expressions." +++ - -Boolean expression matching is a core operation in content-based publish/subscribe -systems. The BE-Tree is a tree-based index structure that efficiently matches large -sets of Boolean expressions over high-dimensional discrete attribute spaces. The -companion BEGen workload generator produces the Boolean-expression workloads used -to evaluate the index under controllable characteristics. diff --git a/content/publications/benchmarking-a-car-originated-signal-approach-for-real-time-electric-vehicle-charging-control.md b/content/publications/benchmarking-a-car-originated-signal-approach-for-real-time-electric-vehicle-charging-control.md new file mode 100644 index 0000000..1abc919 --- /dev/null +++ b/content/publications/benchmarking-a-car-originated-signal-approach-for-real-time-electric-vehicle-charging-control.md @@ -0,0 +1,10 @@ ++++ +title = "Benchmarking a car-originated-signal approach for real-time electric vehicle charging control" +year = 2014 +authors = ["Victor del Razo", "Christoph Goebel", "Hans-Arno Jacobsen"] +venue = "ISGT 2014" +publication_type = "Conference Paper" +research = [] +external_url = "https://doi.org/10.1109/isgt.2014.6816373" +abstract = "We propose and benchmark a new approach for electric vehicle charging control to match arbitrary power profiles. The approach enables vehicles to compute signals reflecting their need for charge and willingness to supply power. An aggregator collects these signals and implements the control with minor computational effort. We benchmark our approach against a centralized optimization, focusing on the trade-off between objective fulfillment and solving time. For the evaluation, we aim at load leveling in a distribution network with high amounts of solar generation. The scenario is based on electricity demand, solar generation, and car mobility data from Munich, Germany. Our results show that the proposed approach achieves relatively good performance, even for large EV fleets, at a low computational cost. Our approach can be generalized to different loads and objectives and could enable new business models for aggregators." ++++ diff --git a/content/publications/benchmarking-apache-kafka-under-network-faults.md b/content/publications/benchmarking-apache-kafka-under-network-faults.md index ffe5c82..921dc17 100644 --- a/content/publications/benchmarking-apache-kafka-under-network-faults.md +++ b/content/publications/benchmarking-apache-kafka-under-network-faults.md @@ -1,11 +1,11 @@ +++ -title = "Benchmarking apache kafka under network faults" +title = "Benchmarking Apache Kafka under network faults" year = 2021 authors = ["Murtaza Raza", "Jawad Tahir", "Christoph Doblander", "Ruben Mayer", "Hans-Arno Jacobsen"] venue = "Proceedings of the 22nd International Middleware Conference: Demos and Posters" publication_type = "Conference Paper" research = ["data-management"] tags = ["benchmarking"] -external_url = "https://dl.acm.org/doi/10.1145/3491086.3492470" -source_url = "https://msrg.utoronto.ca/publications/?page=2" +external_url = "https://doi.org/10.1145/3491086.3492470" +abstract = "Network faults are often transient and hence hard to detect and difficult to resolve. Our study conducts an analysis of Kafka's network fault tolerance capabilities, one of the widely used distributed stream processing system (DSPS). Across different Kafka configurations, we observed that Kafka is fault-tolerant towards network faults to some degree, and we report observations of its shortcomings. We also define a network fault-tolerance benchmark on which other DSPSs can be evaluated." +++ diff --git a/content/publications/bert4nilm-a-bidirectional-transformer-model-for-non-intrusive-load-monitoring.md b/content/publications/bert4nilm-a-bidirectional-transformer-model-for-non-intrusive-load-monitoring.md new file mode 100644 index 0000000..00862d4 --- /dev/null +++ b/content/publications/bert4nilm-a-bidirectional-transformer-model-for-non-intrusive-load-monitoring.md @@ -0,0 +1,10 @@ ++++ +title = "BERT4NILM: A Bidirectional Transformer Model for Non-Intrusive Load Monitoring" +year = 2020 +authors = ["Zhenrui Yue", "Camilo Requena Witzig", "Daniel Jorde", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 5th International Workshop on Non-Intrusive Load Monitoring" +publication_type = "Conference Paper" +research = ["distributed-machine-learning"] +external_url = "https://doi.org/10.1145/3427771.3429390" +abstract = "Non-intrusive load monitoring (NILM) based energy disaggregation is the decomposition of a system's energy into the consumption of its individual appliances. Previous work on deep learning NILM algorithms has shown great potential in the field of energy management and smart grids. In this paper, we propose BERT4NILM, an architecture based on bidirectional encoder representations from transformers (BERT) and an improved objective function designed specifically for NILM learning. We adapt the bidirectional transformer architecture to the field of energy disaggregation and follow the pattern of sequence-to-sequence learning. With the improved loss function and masked training, BERT4NILM outperforms state-of-the-art models across various metrics on the two publicly available datasets UK-DALE and REDD." ++++ diff --git a/content/publications/beyond-message-passing-modern-gnn-architectures-for-online-planner-selection.md b/content/publications/beyond-message-passing-modern-gnn-architectures-for-online-planner-selection.md new file mode 100644 index 0000000..591a9f3 --- /dev/null +++ b/content/publications/beyond-message-passing-modern-gnn-architectures-for-online-planner-selection.md @@ -0,0 +1,10 @@ ++++ +title = "Beyond Message Passing: Modern GNN Architectures for Online Planner Selection" +year = 2026 +authors = ["Jana Vatter", "Ruben Mayer", "Hans-Arno Jacobsen", "Horst Samulowitz", "Michael Katz"] +venue = "Proceedings of the International Conference on Automated Planning and Scheduling" +publication_type = "Conference Paper" +research = ["distributed-machine-learning"] +external_url = "https://doi.org/10.1609/icaps.v36i1.42845" +abstract = "As planning is computationally hard, the performance of automated planners varies greatly across planning tasks. Thus, the ability to predict planner performance on a given task is of great importance. While various learning methods have been applied in cost-optimal planning, Graph Neural Networks (GNNs) were found to perform well. However, existing work only explores a limited range of homogeneous GNN architectures and focuses primarily on the model perspective. We address these limitations by approaching the problem from both data and model perspectives. From the data perspective, we analyze the planners' performance data in terms of Shapley values to assess the potential contribution of individual planners to a portfolio. Our insights enable us to effectively reduce the portfolio from 17 to 6 planners, improving practicality and performance. From the model perspective, our work extends previous investigations of homogeneous graphs by modeling planning tasks as heterogeneous graphs and applying the heterogeneous Relational Graph Convolutional Network (RGCN) and Relational Graph Attention Network (RGAT) models. To analyze the problem in more depth, we thoroughly investigate the impact of GNN model, graph representation, node features, and prediction task. Going further, we propose a hybrid approach in which graph representations obtained by GNNs are used as input to a classical machine learning model (XGBoost), resulting in both a more resource-efficient and accurate approach. Our best model (RGCN+XGBoost) achieves 91.7% accuracy, a substantial improvement over previous methods with 87%, while requiring fewer computational resources. Overall, we demonstrate the effectiveness of heterogeneous GNN-based online planner selection methods, opening up new exciting avenues for future research." ++++ diff --git a/content/publications/beyond-performance-measuring-the-environmental-impact-of-analytical-databases.md b/content/publications/beyond-performance-measuring-the-environmental-impact-of-analytical-databases.md new file mode 100644 index 0000000..b0f3a62 --- /dev/null +++ b/content/publications/beyond-performance-measuring-the-environmental-impact-of-analytical-databases.md @@ -0,0 +1,10 @@ ++++ +title = "Beyond Performance: Measuring the Environmental Impact of Analytical Databases" +year = 2025 +authors = ["Michail Bachras", "Hans-Arno Jacobsen"] +venue = "arXiv" +publication_type = "ArXiv Preprint" +research = ["data-management"] +external_url = "https://arxiv.org/abs/2504.18980" +abstract = "The exponential growth of data is making query processing increasingly critical for modern computing infrastructure, yet the environmental impact of database operations remains poorly understood and largely overlooked. This paper presents ATLAS, a comprehensive methodology for measuring and quantifying the environmental footprint of analytical database systems, considering both operational impacts and manufacturing costs of hardware components. Through extensive empirical evaluation of four distinct database architectures (DuckDB, MonetDB, Hyper, and StarRocks), we uncover how fundamental architectural decisions affect environmental efficiency. Our findings reveal that environmental considerations in database operations are multifaceted, encompassing both immediate operational impacts and long-term sustainability implications. We demonstrate that architectural choices can significantly influence both power consumption and environmental sustainability, while deployment location emerges as a critical factor that can amplify or diminish these architectural advantages." ++++ diff --git a/content/publications/big-data-generation.md b/content/publications/big-data-generation.md new file mode 100644 index 0000000..79628bb --- /dev/null +++ b/content/publications/big-data-generation.md @@ -0,0 +1,9 @@ ++++ +title = "Big Data Generation" +year = 2014 +authors = ["Tilmann Rabl", "Hans-Arno Jacobsen"] +venue = "Lecture Notes in Computer Science; Specifying Big Data Benchmarks" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1007/978-3-642-53974-9_3" ++++ diff --git a/content/publications/bigbench-specification-v0-1-bigbench-an-industry-standard-benchmark-for-big-data-analytics.md b/content/publications/bigbench-specification-v0-1-bigbench-an-industry-standard-benchmark-for-big-data-analytics.md new file mode 100644 index 0000000..a3599b7 --- /dev/null +++ b/content/publications/bigbench-specification-v0-1-bigbench-an-industry-standard-benchmark-for-big-data-analytics.md @@ -0,0 +1,9 @@ ++++ +title = "BigBench Specification V0.1 - BigBench: An Industry Standard Benchmark for Big Data Analytics" +year = 2014 +authors = ["Tilmann Rabl", "Ahmad Ghazal", "Minqing Hu", "Alain Crolotte", "Francois Raab", "Meikel Poess", "Hans-Arno Jacobsen"] +venue = "Lecture Notes in Computer Science; Specifying Big Data Benchmarks" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1007/978-3-642-53974-9_14" ++++ diff --git a/content/publications/bigbench-towards-an-industry-standard-benchmark-for-big-data-analytics.md b/content/publications/bigbench-towards-an-industry-standard-benchmark-for-big-data-analytics.md new file mode 100644 index 0000000..29fafaa --- /dev/null +++ b/content/publications/bigbench-towards-an-industry-standard-benchmark-for-big-data-analytics.md @@ -0,0 +1,10 @@ ++++ +title = "BigBench: towards an industry standard benchmark for big data analytics" +year = 2013 +authors = ["Ahmad Ghazal", "Tilmann Rabl", "Minqing Hu", "Francois Raab", "Meikel Poess", "Alain Crolotte", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 2013 ACM SIGMOD International Conference on Management of Data" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/2463676.2463712" +abstract = "There is a tremendous interest in big data by academia, industry and a large user base. Several commercial and open source providers unleashed a variety of products to support big data storage and processing. As these products mature, there is a need to evaluate and compare the performance of these systems." ++++ diff --git a/content/publications/blockaim-a-neural-network-based-intelligent-middleware-for-large-scale-iot-data-placement-decisions.md b/content/publications/blockaim-a-neural-network-based-intelligent-middleware-for-large-scale-iot-data-placement-decisions.md new file mode 100644 index 0000000..f53efc3 --- /dev/null +++ b/content/publications/blockaim-a-neural-network-based-intelligent-middleware-for-large-scale-iot-data-placement-decisions.md @@ -0,0 +1,10 @@ ++++ +title = "BlockAIM: A Neural Network-Based Intelligent Middleware For Large-Scale IoT Data Placement Decisions" +year = 2023 +authors = ["Syed Muhammad Danish", "Kaiwen Zhang", "Hans-Arno Jacobsen"] +venue = "IEEE Transactions on Mobile Computing" +publication_type = "Journal Article" +research = ["distributed-machine-learning", "data-management"] +external_url = "https://doi.org/10.1109/tmc.2021.3071576" +abstract = "Current Internet of Things (IoT) infrastructures rely on cloud storage however, relying on a single cloud provider puts limitations on the IoT applications and Service Level Agreement (SLA) requirements. Recently, multiple decentralized storage solutions (e.g., based on blockchains) have entered the market with distinct architecture, Quality of Service (QoS) parameters and at lower price compared to the cloud storage. In this work, we introduce BAM: a neural network-based middleware designed for intelligent selection of storage technology for IoT applications. We first propose a blockchain-based data placement protocol and theoretically model a decision optimization problem, which jointly considers cloud, multi-cloud and decentralized storage technologies to select the appropriate medium to store large-scale IoT data, while ensuring data integrity, traceability, auditability and decision verifiability. We then propose a neural network-based maintenance reconfiguration, which aims to optimize the computational complexity of the middleware design along with the blockchain transaction and storage overhead by learning and predicting the applications parameters. We also propose the aggregation rate feedback functionality in our design and model it as a linear optimization problem to improve data quality and precision. Finally, we provide a reference implementation and perform extensive experiments, which demonstrate the effectiveness of the proposed design." ++++ diff --git a/content/publications/blockam-an-adaptive-middleware-for-intelligent-data-storage-selection-for-internet-of-things.md b/content/publications/blockam-an-adaptive-middleware-for-intelligent-data-storage-selection-for-internet-of-things.md new file mode 100644 index 0000000..d6e28a4 --- /dev/null +++ b/content/publications/blockam-an-adaptive-middleware-for-intelligent-data-storage-selection-for-internet-of-things.md @@ -0,0 +1,10 @@ ++++ +title = "BlockAM: An Adaptive Middleware for Intelligent Data Storage Selection for Internet of Things" +year = 2020 +authors = ["Syed Muhammad Danish", "Kaiwen Zhang", "Hans-Arno Jacobsen"] +venue = "2020 IEEE International Conference on Decentralized Applications and Infrastructures (DAPPS)" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/dapps49028.2020.00007" +abstract = "Current Internet of Things (IoT) infrastructures, with its massive data requirements, rely on cloud storage: however, usage of a single cloud storage can place limitations on the IoT applications in terms of service requirements (performance, availability, security etc.). Multi-cloud storage architecture has been emerged as a promising infrastructure to solve this problem, but this approach has limited impact due to the lack of differentiation between competing cloud solutions. Multiple decentralized storage solutions (e.g., based on blockchains) are entering the market with distinct characteristics in terms of architecture, performance, security and availability and at a lower price compared to cloud storage. In this work, we introduce BlockAM: an adaptive middleware for the intelligent selection of storage technology for IoT applications, which jointly considers the cloud, multi-cloud and decentralized storage technologies to store large-scale IoT data. We model the cost-minimization storage selection problem and propose two heuristic algorithms: Dynamic Programming (DP) based algorithm and Greedy Style (GS) algorithm, for optimizing the choice of data storage based on IoT application's service requirements. We also employ blockchain to store IoT data on-chain in order to provide data integrity, auditability and accountability to the middleware architecture. Comparisons among the heuristic algorithms are conducted through extensive experiments, which demonstrates that DP heuristic and GS heuristic achieve up to 92% and 80% accuracy respectively. Moreover, the price associated with a specific IoT application data storage decrease by up to 31.2% by employing our middleware solution." ++++ diff --git a/content/publications/blockchain-for-v2x-a-taxonomy-of-design-use-cases-and-system-requirements.md b/content/publications/blockchain-for-v2x-a-taxonomy-of-design-use-cases-and-system-requirements.md index 8059474..9745e90 100644 --- a/content/publications/blockchain-for-v2x-a-taxonomy-of-design-use-cases-and-system-requirements.md +++ b/content/publications/blockchain-for-v2x-a-taxonomy-of-design-use-cases-and-system-requirements.md @@ -1,11 +1,11 @@ +++ -title = "Blockchain for V2X: A taxonomy of design use cases and system requirements" +title = "Blockchain for V2X: A Taxonomy of Design Use Cases and System Requirements" year = 2021 -authors = ["James Meijers", "Edward Au", "Yuxi Cai", "Hans-Arno Jacobsen", "Shashank Motepalli", "Robert Sun", "Andreas Veneris", "Gengrui Zhang", "Shiquan Zhang"] +authors = ["James Meijers", "Edward Au", "Yuxi Cai", "Hans-Arno Jacobsen", "Shashank Motepalli", "Robert Sun", "Andreas G. Veneris", "Gengrui Zhang", "Shiquan Zhang"] venue = "2021 3rd Conference on Blockchain Research & Applications for Innovative Networks and Services (BRAINS)" publication_type = "Conference Paper" research = ["data-management"] tags = ["blockchain"] -external_url = "https://ieeexplore.ieee.org/document/9569796/" -source_url = "https://msrg.utoronto.ca/publications/?page=2" +external_url = "https://doi.org/10.1109/brains52497.2021.9569796" +abstract = "Vehicles today contain a multitude of sensors creating vast amounts of data. For many applications, these data need to be shared with other entities so that they can also utilize it. Vehicle-to-everything (V2X) is the amalgamation of all potential vehicle communication systems. V2X technologies are enabling many smart-vehicle applications, such as autonomous vehicles. However, in utilizing these data from external entities, vehicles rely on the availability and trustworthiness of centralized entities who may be able to delete, forge, leak, or otherwise tamper with the underlying data. Blockchain technology provides a decentralized mechanism to allow vehicles to validate data they receive in a trustless manner. This paper explores potential applications of blockchain technology in the V2X space, categorizing and analyzing use cases based on their underlying blockchain requirements. It then uses this analysis to determine the key requirements behind an effective V2X blockchain." +++ diff --git a/content/publications/blockchain-for-v2x-applications-and-architectures.md b/content/publications/blockchain-for-v2x-applications-and-architectures.md index ceddb70..4bcbbae 100644 --- a/content/publications/blockchain-for-v2x-applications-and-architectures.md +++ b/content/publications/blockchain-for-v2x-applications-and-architectures.md @@ -1,11 +1,11 @@ +++ -title = "Blockchain for v2x: Applications and architectures" +title = "Blockchain for V2X: Applications and Architectures" year = 2022 authors = ["James Meijers", "Panagiotis Michalopoulos", "Shashank Motepalli", "Gengrui Zhang", "Shiquan Zhang", "Andreas Veneris", "Hans-Arno Jacobsen"] venue = "IEEE Open Journal of Vehicular Technology" publication_type = "Journal Article" research = ["data-management"] tags = ["blockchain"] -external_url = "https://ieeexplore.ieee.org/document/9769951/" -source_url = "https://msrg.utoronto.ca/publications/?page=2" +external_url = "https://doi.org/10.1109/ojvt.2022.3172709" +abstract = "Modern vehicles rely on data from a vast array of sensors such as radar and GPS equipment that can be shared with surrounding vehicles and other interested parties. Vehicle-to-Everything (V2X) is the collection of systems that enable such communication. Although this data sharing has the potential to improve both the safety and efficiency of vehicles, ensuring that what is shared has not been altered, deleted, forged, leaked, or otherwise tampered with remains a challenging problem. Today, blockchain technology allows a system's participants to come to an agreement (consensus) on the state of the system and its data in a decentralized, trustless manner. This new technology may be capable of securing V2X data, as well as enabling other useful V2X services such as payments. However, the V2X ecosystem poses several unique challenges that complicate the application of blockchain technology, not least of which is the vast number of communications that any proposed blockchain network will need to support. This paper gives an overview of V2X and blockchain technology, explores potential applications of blockchain within the V2X domain and justifies its importance. It also reviews, analyzes, and discusses various blockchain architectures that could support V2X applications." +++ diff --git a/content/publications/blockev-efficient-and-secure-charging-station-selection-for-electric-vehicles.md b/content/publications/blockev-efficient-and-secure-charging-station-selection-for-electric-vehicles.md new file mode 100644 index 0000000..1a072e3 --- /dev/null +++ b/content/publications/blockev-efficient-and-secure-charging-station-selection-for-electric-vehicles.md @@ -0,0 +1,10 @@ ++++ +title = "BlockEV: Efficient and Secure Charging Station Selection for Electric Vehicles" +year = 2021 +authors = ["Syed Muhammad Danish", "Kaiwen Zhang", "Hans-Arno Jacobsen", "Nouman Ashraf", "Hassaan Khaliq Qureshi"] +venue = "IEEE Transactions on Intelligent Transportation Systems" +publication_type = "Journal Article" +research = [] +external_url = "https://doi.org/10.1109/tits.2020.3044890" +abstract = "The Intelligent Transportation System (ITS) has become essential for the economical and technological development of a country. The maturity of communication technologies (Vehicle to Infrastructure (V2I) and Vehicle to Vehicle (V2V)) and the amalgamation of smart grids, electric vehicles (EVs) and energy trading resulted in a storm of research opportunities for green ITS. In addition, the combination of vehicular communication technologies and ITS enable efficient selection of EV charging stations (CS) and scheduling EVs charging requirements in real-time. However, the untrusted centralized nature of energy markets and EV charging infrastructures result in several privacy and security threats to EV user's private information. These security and privacy threats include targeted advertisements, privacy leakage, selling data to third party, etc. In this work, we propose BlockEV, a blockchain-based efficient CS selection protocol for EVs to ensure the security and privacy of the EV users, availability of the reserved time slots at CSs, high Quality of Service (QoS) and enhanced EV user comfort. First, a blockchain-based framework is introduced to implement secure charging services and trusted reservation for EVs with the execution of smart contract. Second, we focus on the efficient CS selection and propose a mechanism for EVs to select the CS locally without sharing private information to CS, while fulfilling their service requirements. Evaluations show that the proposed BlockEV is scalable with significantly low blockchain transaction and storage overhead." ++++ diff --git a/content/publications/bpm-in-cloud-architectures-business-process-management-with-slas-and-events.md b/content/publications/bpm-in-cloud-architectures-business-process-management-with-slas-and-events.md new file mode 100644 index 0000000..ada8219 --- /dev/null +++ b/content/publications/bpm-in-cloud-architectures-business-process-management-with-slas-and-events.md @@ -0,0 +1,9 @@ ++++ +title = "BPM in Cloud Architectures: Business Process Management with SLAs and Events" +year = 2010 +authors = ["Vinod Muthusamy", "Hans-Arno Jacobsen"] +venue = "Lecture Notes in Computer Science; Business Process Management" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1007/978-3-642-15618-2_2" ++++ diff --git a/content/publications/brief-announcement-constructing-fault-tolerant-overlay-networks-for-topic-based-publish-subscribe.md b/content/publications/brief-announcement-constructing-fault-tolerant-overlay-networks-for-topic-based-publish-subscribe.md new file mode 100644 index 0000000..d8043f3 --- /dev/null +++ b/content/publications/brief-announcement-constructing-fault-tolerant-overlay-networks-for-topic-based-publish-subscribe.md @@ -0,0 +1,10 @@ ++++ +title = "Brief announcement: constructing fault-tolerant overlay networks for topic-based publish/subscribe" +year = 2013 +authors = ["Chen Chen", "Roman Vitenberg", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 2013 ACM symposium on Principles of distributed computing" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/2484239.2484282" +abstract = "We incorporate fault tolerance in designing reliable and scalable overlay networks to support topic-based pub/sub communication. We propose the MinAvg- kTCO problem parameterized by k: use the minimum number of edges to create a k-topic-connected overlay (kTCO) for pub/sub systems, i.e., for each topic the sub-overlay induced by nodes interested in the topic is k-connected." ++++ diff --git a/content/publications/bringing-distributed-energy-storage-to-market.md b/content/publications/bringing-distributed-energy-storage-to-market.md new file mode 100644 index 0000000..cc0a484 --- /dev/null +++ b/content/publications/bringing-distributed-energy-storage-to-market.md @@ -0,0 +1,10 @@ ++++ +title = "Bringing Distributed Energy Storage to Market" +year = 2016 +authors = ["Christoph Goebel", "Hans-Arno Jacobsen"] +venue = "IEEE Transactions on Power Systems" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.1109/tpwrs.2015.2390263" +abstract = "Spatially distributed energy storage devices can provide additional flexibility to system operators, which is needed to transition from primarily fossil fuel based electricity generation to variable renewable generation. Aggregators in charge of controlling distributed energy storage can take advantage of existing economic incentives for more flexibility. However, controlling large numbers of energy storage devices with individual constraints in accordance with the strict rules of existing energy and reserve markets is challenging. The purpose of this paper is to investigate the design and performance of a system that enables aggregators to bring large numbers of dedicated and fully controllable energy storage devices to multiple markets concurrently. In particular, we propose algorithms and heuristic optimization methods that allow aggregators to control such energy resources in accordance with arbitrary market rules and participation strategies. Our evaluation is based on a realistic dual market (reserve and intra-day energy) use case. We find that effective market-conform control of large numbers of energy storage devices using the proposed algorithms is feasible, even on short time scales. Furthermore, our results also indicate that the scalability of the proposed system design can be further improved via parallelization without limiting the reserve/energy brought to market." ++++ diff --git a/content/publications/building-content-based-publish-subscribe-with-dhts.md b/content/publications/building-content-based-publish-subscribe-with-dhts.md index a4f2899..d7db62f 100644 --- a/content/publications/building-content-based-publish-subscribe-with-dhts.md +++ b/content/publications/building-content-based-publish-subscribe-with-dhts.md @@ -1,18 +1,13 @@ +++ title = "Building Content-Based Publish/Subscribe Systems with Distributed Hash Tables" slug = "building-content-based-publish-subscribe-with-dhts" -year = 2003 -authors = ["Duc Tam", "Reza Azimi", "Hans-Arno Jacobsen"] -venue = "International Workshop on Distributed Event-Based Systems (DEBS)" -publication_type = "Workshop Paper" +year = 2004 +authors = ["David K. Tam", "Reza Azimi", "Hans-Arno Jacobsen"] +venue = "DBISP2P" +publication_type = "Conference Paper" research = ["data-management"] tags = ["publish-subscribe", "peer-to-peer", "distributed-hash-table"] -summary = "Describes P2P-ToPSS, a content-based publish/subscribe system built on top of distributed hash tables for peer-to-peer environments, including the workload material used to evaluate subscription coverage and routing." -external_url = "https://link.springer.com/chapter/10.1007/978-3-540-24629-9_11" +external_url = "https://doi.org/10.1007/978-3-540-24629-9_11" related_datasets = ["p2ptopss-workload"] +abstract = "Building distributed content–based publish/subscribe systems has remained a challenge. Existing solutions typically use a relatively small set of trusted computers as brokers, which may lead to scalability concerns for large Internet–scale workloads. Moreover, since each broker maintains state for a large number of users, it may be difficult to tolerate faults at each broker. In this paper we propose an approach to building content–based publish/subscribe systems on top of distributed hash table (DHT) systems. DHT systems have been effectively used for scalable and fault–tolerant resource lookup in large peer–to–peer networks. Our approach provides predicate–based query semantics and supports constrained range queries. Experimental evaluation shows that our approach is scalable to thousands of brokers, although proper tuning is required." +++ - -P2P-ToPSS extends the Toronto Publish/Subscribe System (ToPSS) to peer-to-peer -environments by building content-based routing on top of distributed hash tables. -The paper presents the routing algorithms, subscription model, and the workload -package used to evaluate the system at scale. diff --git a/content/publications/building-fault-tolerant-overlays-with-low-node-degrees-for-topic-based-publish-subscribe.md b/content/publications/building-fault-tolerant-overlays-with-low-node-degrees-for-topic-based-publish-subscribe.md new file mode 100644 index 0000000..32e385a --- /dev/null +++ b/content/publications/building-fault-tolerant-overlays-with-low-node-degrees-for-topic-based-publish-subscribe.md @@ -0,0 +1,10 @@ ++++ +title = "Building Fault-Tolerant Overlays With Low Node Degrees for Topic-Based Publish/Subscribe" +year = 2022 +authors = ["Chen Chen", "Roman Vitenberg", "Hans-Arno Jacobsen"] +venue = "IEEE Transactions on Dependable and Secure Computing" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.1109/tdsc.2021.3080281" +abstract = "We present a new approach for designing reliable and scalable overlay networks to support topic-based pub/sub communication. We propose the ${{\\mathsf {MinAvg}}-{k}{\\mathsf {TCO}}}$ problem parameterized by ${k}$: use the minimum number of edges to create a ${k}$ k-topic-connected overlay(${{k}TCO}$) for pub/sub systems, i.e., for each topic, the sub-overlay induced by nodes interested in the topic is ${k}$-connected. We prove the NP-completeness of ${{\\mathsf {MinAvg}}-{k}{\\mathsf {TCO}}}$ and show a lower-bound for the hardness of its approximation. For ${{\\mathsf {MinAvg}}-{2}{\\mathsf {TCO}}}$, we present GM2, the first polynomial-time algorithm with an approximation ratio. For ${{\\mathsf {MinAvg}}-{k}{\\mathsf {TCO}}}$, where ${k} \\geq {2}$, we propose HararyPT, a simple and efficient heuristic that aligns nodes across different sub-overlays. We experimentally demonstrate the scalability of GM2 and HararyPT with regards to overlay quality under representative pub/sub workloads. GM2 outputs ${{2}TCO}$ with an empirically insignificant increase in the average node degree, e.g., an increase by 4 in a 1000-node network, as compared to the baseline ${{1}TCO}$ produced by the best-known algorithm. Moreover, GM2 reduces the topic diameters by around 50 percent with respect to those in ${{1}TCO}$." ++++ diff --git a/content/publications/cabinet-dynamically-weighted-consensus-made-fast.md b/content/publications/cabinet-dynamically-weighted-consensus-made-fast.md new file mode 100644 index 0000000..4d44a23 --- /dev/null +++ b/content/publications/cabinet-dynamically-weighted-consensus-made-fast.md @@ -0,0 +1,10 @@ ++++ +title = "Cabinet: Dynamically Weighted Consensus Made Fast" +year = 2025 +authors = ["Gengrui Zhang", "Shiquan Zhang", "Michail Bachras", "Yuqiu Zhang", "Hans-Arno Jacobsen"] +venue = "Proceedings of the VLDB Endowment" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.14778/3718057.3718071" +abstract = "Conventional consensus algorithms, such as Paxos and Raft, encounter inefficiencies when applied to large-scale distributed systems due to the requirement of waiting for replies from a majority of nodes. To address these challenges, we propose Cabinet, a novel consensus algorithm that introduces dynamically weighted consensus, allocating distinct weights to nodes based on any given failure thresholds. Cabinet dynamically adjusts nodes' weights according to their responsiveness, assigning higher weights to faster nodes. The dynamic weight assignment maintains an optimal system performance, especially in large-scale and heterogeneous systems where node responsiveness varies. We evaluate Cabinet against Raft with distributed MongoDB and PostgreSQL databases using YCSB and TPC-C workloads. The evaluation results show that Cabinet outperforms Raft in throughput and latency under increasing system scales, complex networks, and failures in both homogeneous and heterogeneous clusters, offering a promising high-performance consensus solution." ++++ diff --git a/content/publications/caching-in-video-cdns-building-strong-lines-of-defense.md b/content/publications/caching-in-video-cdns-building-strong-lines-of-defense.md new file mode 100644 index 0000000..e598f46 --- /dev/null +++ b/content/publications/caching-in-video-cdns-building-strong-lines-of-defense.md @@ -0,0 +1,10 @@ ++++ +title = "Caching in video CDNs: building strong lines of defense" +year = 2014 +authors = ["Kianoosh Mokhtarian", "Hans-Arno Jacobsen"] +venue = "Proceedings of the Ninth European Conference on Computer Systems" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/2592798.2592817" +abstract = "Planet-scale video Content Delivery Networks (CDNs) deliver a significant fraction of the entire Internet traffic. Effective caching at the edge is vital for the feasibility of these CDNs, which can otherwise incur significant monetary costs and resource overloads in the Internet." ++++ diff --git a/content/publications/can-graph-reordering-speed-up-graph-neural-network-training.md b/content/publications/can-graph-reordering-speed-up-graph-neural-network-training.md index 7ac191c..d52cc25 100644 --- a/content/publications/can-graph-reordering-speed-up-graph-neural-network-training.md +++ b/content/publications/can-graph-reordering-speed-up-graph-neural-network-training.md @@ -2,19 +2,10 @@ title = "Can Graph Reordering Speed Up Graph Neural Network Training? An Experimental Study" year = 2024 authors = ["Nikolai Merkel", "Pierre Toussing", "Ruben Mayer", "Hans-Arno Jacobsen"] -venue = "arXiv Preprint" -publication_type = "ArXiv Preprint" +venue = "Proceedings of the VLDB Endowment" +publication_type = "Journal Article" research = ["distributed-machine-learning"] tags = ["graph-systems", "performance", "distributed-training"] -summary = "Experimental study of how graph reordering strategies affect GNN training on CPU and GPU systems." -external_url = "https://arxiv.org/abs/2409.11129" +external_url = "https://doi.org/10.14778/3705829.3705846" +abstract = "Graph neural networks (GNNs) are a type of neural network capable of learning on graph-structured data. However, training GNNs on large-scale graphs is challenging due to iterative aggregations of high-dimensional features from neighboring vertices within sparse graph structures combined with neural network operations. The sparsity of graphs frequently results in suboptimal memory access patterns and longer training time. Graph reordering is an optimization strategy aiming to improve the graph data layout. It has shown to be effective to speed up graph analytics workloads, but its effect on the performance of GNN training has not been investigated yet. The generalization of reordering to GNN performance is nontrivial, as multiple aspects must be considered: GNN hyper-parameters such as the number of layers, the number of hidden dimensions, and the feature size used in the GNN model, neural network operations, large intermediate vertex states, and GPU acceleration. In our work, we close this gap by performing an empirical evaluation of 12 reordering strategies in two state-of-the-art GNN systems, PyTorch Geometric and Deep Graph Library. Our results show that graph reordering is effective in reducing training time for CPU- and GPU-based training, respectively. Further, we find that GNN hyper-parameters influence the effectiveness of reordering, that reordering metrics play an important role in selecting a reordering strategy, that lightweight reordering performs better for GPU-based than for CPU-based training, and that invested reordering time can in many cases be amortized." +++ - -This paper studies whether graph reordering can reduce the cost of training -graph neural networks. The work compares multiple reordering strategies across -two state-of-the-art GNN systems and looks at how the payoff changes with model -shape, feature size, and accelerator choice. - -The result is a systems-focused view of reordering: it can improve training -time, but the value depends heavily on the workload and on how much time is -spent preparing the graph layout in the first place. diff --git a/content/publications/cassandra-an-ssd-boosted-key-value-store.md b/content/publications/cassandra-an-ssd-boosted-key-value-store.md new file mode 100644 index 0000000..f133075 --- /dev/null +++ b/content/publications/cassandra-an-ssd-boosted-key-value-store.md @@ -0,0 +1,10 @@ ++++ +title = "CaSSanDra: An SSD boosted key-value store" +year = 2014 +authors = ["Prashanth Menon", "Tilmann Rabl", "Mohammad Sadoghi", "Hans-Arno Jacobsen"] +venue = "2014 IEEE 30th International Conference on Data Engineering" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icde.2014.6816732" +abstract = "With the ever growing size and complexity of enterprise systems there is a pressing need for more detailed application performance management. Due to the high data rates, traditional database technology cannot sustain the required performance. Alternatives are the more lightweight and, thus, more performant key-value stores. However, these systems tend to sacrifice read performance in order to obtain the desired write throughput by avoiding random disk access in favor of fast sequential accesses. With the advent of SSDs, built upon the philosophy of no moving parts, the boundary between sequential vs. random access is now becoming blurred. This provides a unique opportunity to extend the storage memory hierarchy using SSDs in key-value stores. In this paper, we extensively evaluate the benefits of using SSDs in commercialized key-value stores. In particular, we investigate the performance of hybrid SSD-HDD systems and demonstrate the benefits of our SSD caching and our novel dynamic schema model." ++++ diff --git a/content/publications/challenger-2-0.md b/content/publications/challenger-2-0.md index 91b12bb..d9bd3cc 100644 --- a/content/publications/challenger-2-0.md +++ b/content/publications/challenger-2-0.md @@ -6,6 +6,6 @@ venue = "Proceedings of the 18th ACM International Conference on Distributed and publication_type = "Conference Paper" research = ["distributed-machine-learning", "data-management"] tags = ["benchmarking", "resilience", "event-processing"] -summary = "Conference paper on resilient, automated deployments for the DEBS Grand Challenge setting." -external_url = "https://dl.acm.org/doi/abs/10.1145/3629104.3666027" +external_url = "https://doi.org/10.1145/3629104.3666027" +abstract = "The DEBS Grand Challenge (GC) is a yearly programming competition organized by the DEBS community. The participants of the GC are provided with a dataset and are required to build a solution generating insights from the data. Participants deploy their solutions on the provided virtual machines (VMs). The dataset is disseminated and the solutions' performance is measured using Challenger, an RPC-based service. Developer surveys show a lower adaption of RPC, which may limit the audience of the GC. Furthermore, provisioning of VMs blocks the compute resources, setting a limit on the number of participants. Lastly, Challenger lacks the functionality to test the fault-tolerance capabilities of the solutions, which is a strict non-functional requirement for the solutions." +++ diff --git a/content/publications/choosing-a-classical-planner-with-graph-neural-networks.md b/content/publications/choosing-a-classical-planner-with-graph-neural-networks.md new file mode 100644 index 0000000..984c0c3 --- /dev/null +++ b/content/publications/choosing-a-classical-planner-with-graph-neural-networks.md @@ -0,0 +1,10 @@ ++++ +title = "Choosing a Classical Planner with Graph Neural Networks" +year = 2024 +authors = ["Jana Vatter", "Ruben Mayer", "Hans-Arno Jacobsen", "Horst Samulowitz", "Michael Katz"] +venue = "arXiv" +publication_type = "ArXiv Preprint" +research = ["distributed-machine-learning"] +external_url = "https://arxiv.org/abs/2402.04874" +abstract = "Online planner selection is the task of choosing a solver out of a predefined set for a given planning problem. As planning is computationally hard, the performance of solvers varies greatly on planning problems. Thus, the ability to predict their performance on a given problem is of great importance. While a variety of learning methods have been employed, for classical cost-optimal planning the prevailing approach uses Graph Neural Networks (GNNs). In this work, we continue the line of work on using GNNs for online planner selection. We perform a thorough investigation of the impact of the chosen GNN model, graph representation and node features, as well as prediction task. Going further, we propose using the graph representation obtained by a GNN as an input to the Extreme Gradient Boosting (XGBoost) model, resulting in a more resource-efficient yet accurate approach. We show the effectiveness of a variety of GNN-based online planner selection methods, opening up new exciting avenues for research on online planner selection." ++++ diff --git a/content/publications/clear-a-circuit-level-electric-appliance-radar-for-the-electric-cabinet.md b/content/publications/clear-a-circuit-level-electric-appliance-radar-for-the-electric-cabinet.md new file mode 100644 index 0000000..e63e56f --- /dev/null +++ b/content/publications/clear-a-circuit-level-electric-appliance-radar-for-the-electric-cabinet.md @@ -0,0 +1,10 @@ ++++ +title = "CLEAR - A circuit level electric appliance radar for the electric cabinet" +year = 2017 +authors = ["Anwar Ul Haq", "Thomas Kriechbaumer", "Matthias Kahl", "Hans-Arno Jacobsen"] +venue = "2017 IEEE International Conference on Industrial Technology (ICIT)" +publication_type = "Conference Paper" +research = [] +external_url = "https://doi.org/10.1109/icit.2017.7915521" +abstract = "In this paper, we propose a cost-effective hardware design for acquiring high-resolution energy data from electricity mains to observe appliance switching transients. By utilizing these distinct appliance switching transients, we are interested to predict all appliances operating in a building and also estimate their power consumption. The proposed design fulfills the data acquisition requirements for effective appliance identification from aggregate load signals. Due to its unique and customizable features, the circuit level electric appliance radar (CLEAR) is capable of operating in multiple environments (residential, commercial, and industrial). CLEAR can simultaneously measure voltage and current data streams from a three-phase system at up to 250 kHz to accurately pinpoint individual appliance switching. The first prototype is installed in an office building to collect office level aggregate data. With this aggregate data, the main challenge lies in detecting and isolating similar appliance switching events from switched-mode power supplies (SMPS) utilized by most office appliances such as laptops, LCDs, and printers." ++++ diff --git a/content/publications/cms-topss-efficient-dissemination-of-rss-documents.md b/content/publications/cms-topss-efficient-dissemination-of-rss-documents.md new file mode 100644 index 0000000..98b2a1e --- /dev/null +++ b/content/publications/cms-topss-efficient-dissemination-of-rss-documents.md @@ -0,0 +1,10 @@ ++++ +title = "CMS-ToPSS: Efficient Dissemination of RSS Documents" +year = 2005 +authors = ["Milenko Petrovic", "Haifeng Liu", "Hans-Arno Jacobsen"] +venue = "VLDB" +publication_type = "Conference Paper" +research = [] +external_url = "https://www.vldb.org/conf/2005/papers/p1279-petrovic.pdf" +abstract = "Recent years have seen a rise in the number of unconventional publishing tools on the Internet. Tools such as wikis, blogs, discussion forums, and web-based content management systems have experienced tremendous rise in popularity and use; primarily because they provide something traditional tools do not: easy of use for non computer-oriented users and they are based on the idea of collaboration. It is estimated, by pewinternet.org, that 32 million people in the US read blogs (which represents 27% of the estimated 120 million US Internet users) while 8 million people have said that they have created blogs." ++++ diff --git a/content/publications/coddora-co2-based-occupancy-detection-model-trained-via-domain-randomization.md b/content/publications/coddora-co2-based-occupancy-detection-model-trained-via-domain-randomization.md new file mode 100644 index 0000000..cc1af17 --- /dev/null +++ b/content/publications/coddora-co2-based-occupancy-detection-model-trained-via-domain-randomization.md @@ -0,0 +1,10 @@ ++++ +title = "Coddora: CO2-Based Occupancy Detection Model Trained via Domain Randomization" +year = 2024 +authors = ["Manuel Weber", "Farzan Banihashemi", "Davor Stjelja", "Peter Mandl", "Ruben Mayer", "Hans-Arno Jacobsen"] +venue = "2024 International Joint Conference on Neural Networks (IJCNN)" +publication_type = "Conference Paper" +research = ["distributed-machine-learning"] +external_url = "https://doi.org/10.1109/ijcnn60899.2024.10650820" +abstract = "Information about human presence in indoor spaces is crucial for building energy optimization. While there has been a considerable amount of research on using neural networks to automatically detect occupancy from CO2 sensors, their application in practice is limited due to the scarcity of labeled training data. In this paper, we propose Coddora, an off-the-shelf deep learning model pretrained on data from randomized room simulations. Coddora enables quick adaptation to real-world rooms, requiring only minimal data collection. Our contribution includes two model variants for application via fine-tuning or zero-shot classifying, as well as the synthetic dataset providing data from simulations with 100,000 room models." ++++ diff --git a/content/publications/community-clustering-for-distributed-publish-subscribe-systems.md b/content/publications/community-clustering-for-distributed-publish-subscribe-systems.md new file mode 100644 index 0000000..9f97e36 --- /dev/null +++ b/content/publications/community-clustering-for-distributed-publish-subscribe-systems.md @@ -0,0 +1,10 @@ ++++ +title = "Community Clustering for Distributed Publish/Subscribe Systems" +year = 2012 +authors = ["Wei Li", "Songlin Hu", "Jintao Li", "Hans-Arno Jacobsen"] +venue = "2012 IEEE International Conference on Cluster Computing" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/cluster.2012.67" +abstract = "Optimized placement of clients in a distributed publish/subscribe system is an important technique to improve overall system efficiency. Current methods, like interest clustering or publisher placement, treat a client as, either a pure publisher, or subscriber, but not as both. Also, the cost of client movement is usually ignored. However, many applications based on publish/subscribe systems model clients as publisher and subscriber at the same time, which breaks the assumptions made by current approaches. Considering the complex dependency among clients, we propose a new community-oriented clustering approach, based on the forming of client clusters that exhibit intense communication relationships, while keeping client movement cost low. The evaluation based on a public data set shows that our method is efficient, adapts to different settings of experimental conditions, and wins over the popular interest clustering approach with respect to number of messages sent, propagation hop count and end-to-end latency." ++++ diff --git a/content/publications/composite-subscriptions-in-content-based-publish-subscribe-systems.md b/content/publications/composite-subscriptions-in-content-based-publish-subscribe-systems.md new file mode 100644 index 0000000..733dcc9 --- /dev/null +++ b/content/publications/composite-subscriptions-in-content-based-publish-subscribe-systems.md @@ -0,0 +1,9 @@ ++++ +title = "Composite Subscriptions in Content-Based Publish/Subscribe Systems" +year = 2005 +authors = ["Guoli Li", "Hans-Arno Jacobsen"] +venue = "Lecture Notes in Computer Science; Middleware 2005" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1007/11587552_13" ++++ diff --git a/content/publications/configurable-hardware-based-streaming-architecture-using-online-programmable-blocks.md b/content/publications/configurable-hardware-based-streaming-architecture-using-online-programmable-blocks.md new file mode 100644 index 0000000..38716a4 --- /dev/null +++ b/content/publications/configurable-hardware-based-streaming-architecture-using-online-programmable-blocks.md @@ -0,0 +1,10 @@ ++++ +title = "Configurable hardware-based streaming architecture using Online Programmable-Blocks" +year = 2015 +authors = ["Mohammadreza Najafi", "Mohammad Sadoghi", "Hans-Arno Jacobsen"] +venue = "2015 IEEE 31st International Conference on Data Engineering" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icde.2015.7113336" +abstract = "The limitations of traditional general-purpose processors have motivated the use of specialized hardware solutions (e.g., FPGAs) to achieve higher performance in stream processing. However, state-of-the-art hardware-only solutions have limited support to adapt to changes in the query workload." ++++ diff --git a/content/publications/congestion-avoidance-with-incremental-filter-aggregation-in-content-based-routing-networks.md b/content/publications/congestion-avoidance-with-incremental-filter-aggregation-in-content-based-routing-networks.md new file mode 100644 index 0000000..4c23685 --- /dev/null +++ b/content/publications/congestion-avoidance-with-incremental-filter-aggregation-in-content-based-routing-networks.md @@ -0,0 +1,10 @@ ++++ +title = "Congestion Avoidance with Incremental Filter Aggregation in Content-Based Routing Networks" +year = 2015 +authors = ["Mingwen Chen", "Songlin Hu", "Vinod Muthusamy", "Hans-Arno Jacobsen"] +venue = "2015 IEEE 35th International Conference on Distributed Computing Systems" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icdcs.2015.63" +abstract = "The subscription covering optimization, whereby a general subscription quenches the forwarding of more specific ones, is a common technique to reduce network traffic and routing state in content-based routing networks. Such optimizations, however, leave the system vulnerable to unsubscriptions that trigger the immediate forwarding of all the subscriptions they had previously quenched. These subscription bursts can severely congest the network, and destabilize the system. This paper presents techniques to retain much of the benefits of subscription covering while avoiding bursty subscription traffic. Heuristics are used to estimate the similarity among subscriptions, and a distributed algorithm determines the portions of a subscription propagation tree that should be preserved. Evaluations show that these mechanisms avoid subscription bursts while maintaining relatively compact routing tables." ++++ diff --git a/content/publications/content-based-routing-in-mobile-ad-hoc-networks.md b/content/publications/content-based-routing-in-mobile-ad-hoc-networks.md new file mode 100644 index 0000000..20bfa00 --- /dev/null +++ b/content/publications/content-based-routing-in-mobile-ad-hoc-networks.md @@ -0,0 +1,10 @@ ++++ +title = "Content-Based Routing in Mobile Ad Hoc Networks" +year = 2005 +authors = ["Milenko Petrovic", "Vinod Muthusamy", "Hans-Arno Jacobsen"] +venue = "The Second Annual International Conference on Mobile and Ubiquitous Systems: Networking and Services" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/mobiquitous.2005.18" +abstract = "The publish/subscribe model of communication provides sender/receiver decoupling and selective information dissemination that is appropriate for mobile environments characterized by scarce resources and a lack of fixed infrastructure. We propose and evaluate three content-based routing protocols: CBR is an adaptation of existing distributed publish/subscribe protocols for wired networks, FT-CBR extends CBR to provide fault-tolerance, and RAFT-CBR provides both fault-tolerance and reliability. Using network simulations we analyze the applicability and test the tradeoffs of these algorithms. We show that RAFT-CBR can guarantee 100% delivery to small groups, at the expense of transmission delay. CBR, with a low message overhead and low delay, is more suitable for larger groups at the expense of reliability. FT-CBR provides comparable delivery rates to RAFT-CBR, as well as low delay, at the expense of increased message cost." ++++ diff --git a/content/publications/coordinated-caching-in-planet-scale-cdns-analysis-of-feasibility-and-benefits.md b/content/publications/coordinated-caching-in-planet-scale-cdns-analysis-of-feasibility-and-benefits.md new file mode 100644 index 0000000..a24c464 --- /dev/null +++ b/content/publications/coordinated-caching-in-planet-scale-cdns-analysis-of-feasibility-and-benefits.md @@ -0,0 +1,10 @@ ++++ +title = "Coordinated caching in planet-scale CDNs: Analysis of feasibility and benefits" +year = 2016 +authors = ["Kianoosh Mokhtarian", "Hans-Arno Jacobsen"] +venue = "IEEE INFOCOM 2016 - The 35th Annual IEEE International Conference on Computer Communications" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/infocom.2016.7524378" +abstract = "Video Content Distribution Networks (CDNs) serve a significant fraction of the Internet traffic through a global network of cache servers. In a planet-scale CDN with millions of videos, cache servers only consider their own request patterns for managing their content. We analyze how, in the absence of cooperative caching, the knowledge of requests in remote serving locations can lead to better caching decisions overall and can reduce serving costs. We call this practice cache coordination. Our analyses in this paper are based on actual video workload data from a global CDN. We analyze the spatial correlation of video popularities worldwide, the effectiveness and feasibility of cache coordination, and its scalability: from a city to across countries." ++++ diff --git a/content/publications/d2worm-a-management-infrastructure-for-distributed-data-centric-workflows.md b/content/publications/d2worm-a-management-infrastructure-for-distributed-data-centric-workflows.md new file mode 100644 index 0000000..4aac1a9 --- /dev/null +++ b/content/publications/d2worm-a-management-infrastructure-for-distributed-data-centric-workflows.md @@ -0,0 +1,10 @@ ++++ +title = "D2WORM: A Management Infrastructure for Distributed Data-centric Workflows" +year = 2015 +authors = ["Martin Jergler", "Mohammad Sadoghi", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 2015 ACM SIGMOD International Conference on Management of Data" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/2723372.2735362" +abstract = "Unlike traditional activity-flow-based models, data-centric workflows primarily focus on the data to drive a business. This enables the unification of operational management, concurrent process analytics, compliance with process or associated data constraints, and adaptability to changing environments. In this demonstration, we present D2Worm, a Distributed Data-centric Workflow Management system. D2Worm allows users to (1) graphically model data-centric workflows in a declarative fashion based on the Guard-Stage-Milestone (GSM) meta-model, (2) automatically compile the modelled workflow into several fine-granular workflow units (WFUs), and (3) deploy these WFUs on distributed infrastructures. A WFU is a system component that manages a subset of the workflow's data model and, at the same time, represents part of the global control flow by evaluating conditions over the data. WFUs communicate with each other over a publish/subscribe messaging infrastructure that allows the architecture to scale from a single node to dozens of machines distributed over different data-centers. In addition, D2Worm is able to (4) concurrently execute multiple workflow instances and monitor their behavior in real-time." ++++ diff --git a/content/publications/debs-2022-grand-challenge-data-set-trading-data.md b/content/publications/debs-2022-grand-challenge-data-set-trading-data.md index 7448718..3f9426a 100644 --- a/content/publications/debs-2022-grand-challenge-data-set-trading-data.md +++ b/content/publications/debs-2022-grand-challenge-data-set-trading-data.md @@ -7,5 +7,4 @@ publication_type = "Dataset" research = ["data-management"] tags = ["benchmarking"] external_url = "https://zenodo.org/records/6382482" -source_url = "https://msrg.utoronto.ca/publications/?page=2" +++ diff --git a/content/publications/decentralization-in-pos-blockchain-consensus-quantification-and-advancement.md b/content/publications/decentralization-in-pos-blockchain-consensus-quantification-and-advancement.md new file mode 100644 index 0000000..c288d14 --- /dev/null +++ b/content/publications/decentralization-in-pos-blockchain-consensus-quantification-and-advancement.md @@ -0,0 +1,10 @@ ++++ +title = "Decentralization in PoS Blockchain Consensus: Quantification and Advancement" +year = 2025 +authors = ["Shashank Motepalli", "Hans-Arno Jacobsen"] +venue = "IEEE Transactions on Network and Service Management" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.1109/tnsm.2025.3561098" +abstract = "Decentralization is a foundational principle of permissionless blockchains, with consensus mechanisms serving a critical role in its realization. This study quantifies the decentralization of consensus mechanisms in proof-of-stake (PoS) blockchains using a comprehensive set of metrics, including Nakamoto coefficients, Gini, Herfindahl-Hirschman Index (HHI), Shapley values, and Zipf’s coefficient. Our empirical analysis across ten prominent blockchains reveals significant concentration of stake among a few validators, posing challenges to fair consensus. To address this, we introduce two alternative weighting models for PoS consensus: Square Root Stake Weight (SRSW) and Logarithmic Stake Weight (LSW), which adjust validator influence through non-linear transformations. Results demonstrate that SRSW and LSW models improve decentralization metrics by an average of 51% and 132%, respectively, supporting more equitable and resilient blockchain systems." ++++ diff --git a/content/publications/decentralized-and-policy-aware-serverless-orchestration-for-the-federated-web.md b/content/publications/decentralized-and-policy-aware-serverless-orchestration-for-the-federated-web.md new file mode 100644 index 0000000..c2ebff7 --- /dev/null +++ b/content/publications/decentralized-and-policy-aware-serverless-orchestration-for-the-federated-web.md @@ -0,0 +1,10 @@ ++++ +title = "Decentralized and Policy-Aware Serverless Orchestration for the Federated Web" +year = 2025 +authors = ["Yuqiu Zhang", "Hans-Arno Jacobsen"] +venue = "Companion Proceedings of the ACM on Web Conference 2025" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/3701716.3715573" +abstract = "As Web services and Web-of-Things (WoT) applications increasingly span multiple administrative domains such as clouds, edges, and community-run platforms, centralized serverless orchestration frameworks struggle to handle heterogeneous policies, data governance constraints, and dynamic workloads. This situation prompts a new design space: federated and decentralized scheduling models for serverless computing that do not rely on a single global coordinator. In this paper, we explore an approach in which each administrative domain autonomously manages its own resources and complies with local regulations, while lightweight decentralized protocols enable these domains to cooperatively share load information, offer resource availability hints, and negotiate function placements. We illustrate how a small set of core mechanisms such as metadata exchange protocols, compliance-driven constraints, and trust-aware placement negotiations, could enable serverless functions to be flexibly placed and migrated across a federated ecosystem. Our experimental analysis suggests that decentralized coordination can maintain service-level objectives (SLOs) under diverse conditions, especially under load spikes, and align better with the policy and compliance requirements that characterize a heterogeneous Web environment. This paper charts a path forward for research on federated serverless orchestration, aiming to foster more flexible, scalable, and policy-aware infrastructure at global scale." ++++ diff --git a/content/publications/decentralized-execution-of-event-driven-scientific-workflows.md b/content/publications/decentralized-execution-of-event-driven-scientific-workflows.md new file mode 100644 index 0000000..d39d287 --- /dev/null +++ b/content/publications/decentralized-execution-of-event-driven-scientific-workflows.md @@ -0,0 +1,10 @@ ++++ +title = "Decentralized Execution of Event-Driven Scientific Workflows" +year = 2006 +authors = ["Guoli Li", "Vinod Muthusamy", "Hans-Arno Jacobsen", "Serge Mankovski"] +venue = "2006 IEEE Services Computing Workshops" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/scw.2006.10" +abstract = "Scientific workflows (SWF) are traditionally coordinated and executed in a centralized fashion. This creates a single point of failure, forms a scalability bottleneck, and often leads to too much message traffic routed back to the coordinator. We have developed PADRES, a content-based publish/subscribe platform that serves as a runtime environment for the decentralized execution, control, and monitoring of SWF. Publish/subscribe is a natural paradigm for event-driven applications such as SWF management, as the loosely-coupled nature of publishers and subscribers relieves the coordinator from maintaining client connection and capability information. PADRES has been developed with features inspired by the requirements of SWF management. Its unique features include an expressive subscription language, composite subscription processing support, a rule-based matching and routing mechanism, a query-based historic data access mechanism, and support for the decentralized execution of SWFs specified in XML" ++++ diff --git a/content/publications/decentralizing-permissioned-blockchain-with-delay-towers.md b/content/publications/decentralizing-permissioned-blockchain-with-delay-towers.md index 3b96d31..8f4ac90 100644 --- a/content/publications/decentralizing-permissioned-blockchain-with-delay-towers.md +++ b/content/publications/decentralizing-permissioned-blockchain-with-delay-towers.md @@ -1,11 +1,11 @@ +++ -title = "Decentralizing permissioned blockchain with delay towers" +title = "Decentralizing Permissioned Blockchain with Delay Towers" year = 2022 authors = ["Shashank Motepalli", "Hans-Arno Jacobsen"] -venue = "arXiv Preprint" +venue = "arXiv" publication_type = "ArXiv Preprint" research = ["data-management"] tags = ["blockchain"] external_url = "https://arxiv.org/abs/2203.09714" -source_url = "https://msrg.utoronto.ca/publications/?page=2" +abstract = "Growing excitement around permissionless blockchains is uncovering its latent scalability concerns. Permissioned blockchains offer high transactional throughput and low latencies while compromising decentralization. In the quest for a decentralized, scalable blockchain fabric, i.e., to offer the scalability of permissioned blockchain in a permissionless setting, we present L4L to encourage decentralization over the permissioned Libra network without compromising its sustainability. L4L employs delay towers, -- puzzle towers that leverage verifiable delay functions -- for establishing identity in a permissionless setting. Delay towers cannot be parallelized due to their sequential execution, making them an eco-friendly alternative. We also discuss methodologies to replace validators participating in consensus to promote compliant behavior. Our evaluations found that the cost of enabling decentralization over permissioned networks is almost negligible. Furthermore, delay towers offer an alternative to existing permissionless consensus mechanisms without requiring airdrops or pre-sale of tokens." +++ diff --git a/content/publications/deconstructing-blockchains-concepts-systems-and-insights.md b/content/publications/deconstructing-blockchains-concepts-systems-and-insights.md new file mode 100644 index 0000000..404fb62 --- /dev/null +++ b/content/publications/deconstructing-blockchains-concepts-systems-and-insights.md @@ -0,0 +1,10 @@ ++++ +title = "Deconstructing Blockchains: Concepts, Systems, and Insights" +year = 2018 +authors = ["Kaiwen Zhang", "Roman Vitenberg", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 12th ACM International Conference on Distributed and Event-based Systems" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/3210284.3219502" +abstract = "Popularly known for powering cryptocurrencies such as Bitcoin and Ethereum, blockchains is seen as a disruptive technology capable of impacting a wide variety of domains, ranging from finance to governance, by offering superior security, reliability, and transparency in a decentralized manner. In this tutorial presentation, we first study the original Bitcoin design, as well as Ethereum and Hyperledger, and reflect on their design from an academic perspective. We provide an overview of potential applications and associated research challenges, as well as a survey of ongoing research projects. We mention opportunities blockchain creates for event-based systems. Finally, we conclude with a walkthrough showing the process of developing a decentralized application (ĐSApp), using a popular Smart Contract language (Solidity) for the blockchain platform of Ethereum." ++++ diff --git a/content/publications/delay-towers-to-bootstrap-blockchains.md b/content/publications/delay-towers-to-bootstrap-blockchains.md index a5b108e..2883aaa 100644 --- a/content/publications/delay-towers-to-bootstrap-blockchains.md +++ b/content/publications/delay-towers-to-bootstrap-blockchains.md @@ -6,6 +6,6 @@ venue = "2024 IEEE International Conference on Blockchain and Cryptocurrency (IC publication_type = "Conference Paper" research = ["data-management"] tags = ["blockchain", "consensus"] -summary = "Conference paper on blockchain bootstrapping and system design for decentralized consensus." -external_url = "https://ieeexplore.ieee.org/abstract/document/10634426/" +external_url = "https://doi.org/10.1109/icbc59979.2024.10634426" +abstract = "To bolster Sybil resistance and establish persistent identities in new permissionless blockchain networks, we introduce delay towers, leveraging Verifiable Delay Functions (VDFs) to implement a proof of elapsed time (PoET). By utilizing inherently sequential cryptographic primitives, delay towers present an eco-friendly alternative to traditional PoW and PoS mechanisms, enabling sustainable network growth without reliance on token sales or airdrops. This work not only considers the blockchain’s ecological footprint but also upholds the principles of a fair launch to enable a decentralized blockchain ecosystem." +++ diff --git a/content/publications/demo-a-framework-for-location-information-processing.md b/content/publications/demo-a-framework-for-location-information-processing.md new file mode 100644 index 0000000..f06392c --- /dev/null +++ b/content/publications/demo-a-framework-for-location-information-processing.md @@ -0,0 +1,10 @@ ++++ +title = "Demo: a framework for location information processing" +year = 2005 +authors = ["Zhengdao Xu", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 6th international conference on Mobile data management" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/1071246.1071301" +abstract = "Recently, with the advances in wireless communications and location positioning technology, the potential for tracking, correlating, and filtering information about moving entities (i.e., generally speaking any moving objects, such as airplanes, automobiles, trucks, cyclists, pedestrians, goods, and packages etc.) has greatly increased. The knowledge of spatial, temporal, and causal relationships between moving objects would allow the support of highly personalized and effective location-based services (LBS). Such services could track, correlate, and process object positions, object profiles, and past, present, and future object movement patterns." ++++ diff --git a/content/publications/dependable-distributed-content-based-publish-subscribe-systems-doctoral-symposium.md b/content/publications/dependable-distributed-content-based-publish-subscribe-systems-doctoral-symposium.md new file mode 100644 index 0000000..403389e --- /dev/null +++ b/content/publications/dependable-distributed-content-based-publish-subscribe-systems-doctoral-symposium.md @@ -0,0 +1,10 @@ ++++ +title = "Dependable distributed content-based publish/subscribe systems: doctoral symposium" +year = 2016 +authors = ["Pooya Salehi", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 10th ACM International Conference on Distributed and Event-based Systems" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/2933267.2933433" +abstract = "Content-based publish/subscribe systems provide an efficient communication paradigm that allows decoupling of information producers and consumers across location and time. Distributed overlay-based publish/subscribe systems, while scalable, face many problems that hinders their applicability in scenarios requiring dependable communication. In this paper, we discuss three important dimensions of dependability in distributed content-based publish/subscribe systems, namely, availability, reliability and maintainability." ++++ diff --git a/content/publications/dgfindex-for-smart-grid-enhancing-hive-with-a-cost-effective-multidimensional-range-index.md b/content/publications/dgfindex-for-smart-grid-enhancing-hive-with-a-cost-effective-multidimensional-range-index.md new file mode 100644 index 0000000..9fb818c --- /dev/null +++ b/content/publications/dgfindex-for-smart-grid-enhancing-hive-with-a-cost-effective-multidimensional-range-index.md @@ -0,0 +1,10 @@ ++++ +title = "DGFIndex for Smart Grid: Enhancing Hive with a Cost-Effective Multidimensional Range Index" +year = 2014 +authors = ["Yue Liu", "Songlin Hu", "Tilmann Rabl", "Wantao Liu", "Hans-Arno Jacobsen", "Kaifeng Wu", "Jian Chen", "Jintao Li"] +venue = "Proceedings of the VLDB Endowment" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.14778/2733004.2733021" +abstract = "In Smart Grid applications, as the number of deployed electric smart meters increases, massive amounts of valuable meter data is generated and collected every day. To enable reliable data collection and make business decisions fast, high throughput storage and high-performance analysis of massive meter data become crucial for grid companies. Considering the advantage of high efficiency, fault tolerance, and price-performance of Hadoop and Hive systems, they are frequently deployed as underlying platform for big data processing. However, in real business use cases, these data analysis applications typically involve multidimensional range queries (MDRQ) as well as batch reading and statistics on the meter data. While Hive is high-performance at complex data batch reading and analysis, it lacks efficient indexing techniques for MDRQ. In this paper, we propose DGFIndex, an index structure for Hive that efficiently supports MDRQ for massive meter data. DGFIndex divides the data space into cubes using the grid file technique. Unlike the existing indexes in Hive, which stores all combinations of multiple dimensions, DGFIndex only stores the information of cubes. This leads to smaller index size and faster query processing. Furthermore, with pre-computing user-defined aggregations of each cube, DGFIndex only needs to access the boundary region for aggregation query. Our comprehensive experiments show that DGFIndex can save significant disk space in comparison with the existing indexes in Hive and the query performance with DGFIndex is 2-50 times faster than existing indexes in Hive and HadoopDB for aggregation query, 2-5 times faster than both for non-aggregation query, 2-75 times faster than scanning the whole table in different query selectivity." ++++ diff --git a/content/publications/diba-a-re-configurable-stream-processor.md b/content/publications/diba-a-re-configurable-stream-processor.md new file mode 100644 index 0000000..15c5777 --- /dev/null +++ b/content/publications/diba-a-re-configurable-stream-processor.md @@ -0,0 +1,10 @@ ++++ +title = "DIBA: A Re-Configurable Stream Processor" +year = 2024 +authors = ["Mohammadreza Najafi", "Thamir M. Qadah", "Mohammad Sadoghi", "Hans-Arno Jacobsen"] +venue = "IEEE Transactions on Knowledge and Data Engineering" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.1109/tkde.2024.3381192" +abstract = "Stream processing acceleration is driven by the continuously increasing volume and velocity of data generated on the Web and the limitations of storage, computation, and power consumption. Hardware solutions provide better performance and power consumption, but they are hindered by the high research and development costs and the long time to market. In this work, we propose our re-configurable stream processor (Diba), a complete rethinking of a previously proposed customized and flexible query processor that targets real-time stream processing. Diba uses a unidirectional dataflow not dedicated to any specific type of query (operator) on streams, allowing a straightforward placement of processing components on a general data path that facilitates query mapping. In Diba, the concepts of the distribution network and processing components are implemented as two separate entities connected using generic interfaces. This approach allows the adoption of a versatile architecture for a family of queries rather than forcing a rigid chain of processing components to implement such queries. Our experimental evaluations of representative queries from TPC-H yielded processing times of 300, 1220, and 3520 milliseconds for data streams with scale factor sizes of one, four, and ten gigabytes, respectively." ++++ diff --git a/content/publications/disconnected-operation-in-publish-subscribe-middleware.md b/content/publications/disconnected-operation-in-publish-subscribe-middleware.md new file mode 100644 index 0000000..971af48 --- /dev/null +++ b/content/publications/disconnected-operation-in-publish-subscribe-middleware.md @@ -0,0 +1,10 @@ ++++ +title = "Disconnected Operation in Publish/Subscribe Middleware" +year = 2004 +authors = ["Ioana Burcea", "Hans-Arno Jacobsen", "Eyal de Lara", "Vinod Muthusamy", "Milenko Petrovic"] +venue = "IEEE International Conference on Mobile Data Management, 2004. Proceedings. 2004" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/mdm.2004.1263041" +abstract = "The decoupling of producers and consumers in time and space in the publish/subscribe paradigm lends itself well to the support of mobile users who roam about the environment and have intermittent network connectivity. This paper identifies the factors that affect the performance of a distributed publish/subscribe architecture supporting mobility; formalizes mobility algorithms for distributed publish/subscribe systems and develops and evaluates optimizations that reduce the costs associated with supporting mobility in publish/subscribe systems. In our analysis, we focus on the \"unicast\" traffic generated to support mobile users, as opposed to the regular \"multicast\" traffic used for event dissemination to stationary clients. We find that the network capacity must be doubled to handle the extra load of just 10% of mobile users." ++++ diff --git a/content/publications/discussion-of-bigbench-a-proposed-industry-standard-performance-benchmark-for-big-data.md b/content/publications/discussion-of-bigbench-a-proposed-industry-standard-performance-benchmark-for-big-data.md new file mode 100644 index 0000000..b2c4a83 --- /dev/null +++ b/content/publications/discussion-of-bigbench-a-proposed-industry-standard-performance-benchmark-for-big-data.md @@ -0,0 +1,9 @@ ++++ +title = "Discussion of BigBench: A Proposed Industry Standard Performance Benchmark for Big Data" +year = 2015 +authors = ["Chaitanya K. Baru", "Milind A. Bhandarkar", "Carlo Curino", "Manuel Danisch", "Michael Frank", "Bhaskar Gowda", "Hans-Arno Jacobsen", "Huang Jie", "Dileep Kumar", "Raghunath Othayoth Nambiar", "Meikel Poess", "Francois Raab", "Tilmann Rabl", "Nishkam Ravi", "Kai Sachs", "Saptak Sen", "Lan Yi", "Choonhan Youn"] +venue = "Lecture Notes in Computer Science; Performance Characterization and Benchmarking. Traditional to Big Data" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1007/978-3-319-15350-6_4" ++++ diff --git a/content/publications/distributed-automatic-service-composition-in-large-scale-systems.md b/content/publications/distributed-automatic-service-composition-in-large-scale-systems.md new file mode 100644 index 0000000..413693b --- /dev/null +++ b/content/publications/distributed-automatic-service-composition-in-large-scale-systems.md @@ -0,0 +1,10 @@ ++++ +title = "Distributed automatic service composition in large-scale systems" +year = 2008 +authors = ["Songlin Hu", "Vinod Muthusamy", "Guoli Li", "Hans-Arno Jacobsen"] +venue = "Proceedings of the second international conference on Distributed event-based systems" +publication_type = "Conference Paper" +research = [] +external_url = "https://doi.org/10.1145/1385989.1386019" +abstract = "Automatic service composition is an active research area in the field of service computing. This paper presents a distributed approach to automatically discover a composition of services based on the desired input to and output from the process. The algorithm makes use of the content-based publish/subscribe model, with service inputs modeled as subscriptions, and outputs as advertisements. Service interfaces are mapped to publish/subscribe messages in such a way that publish/subscribe matching is used to evaluate service compatibility. In this way, large-scale distributed service composition and process discovery is achieved with a distributed publish/subscribe network. Evaluations in a distributed environment of a real implementation of the system demonstrate the scalability of the distributed approach, especially with respect to the number of services, the complexity of the discovered processes, and the number of concurrent searches." ++++ diff --git a/content/publications/distributed-convex-optimization-for-electric-vehicle-aggregators.md b/content/publications/distributed-convex-optimization-for-electric-vehicle-aggregators.md new file mode 100644 index 0000000..0b1ebe6 --- /dev/null +++ b/content/publications/distributed-convex-optimization-for-electric-vehicle-aggregators.md @@ -0,0 +1,10 @@ ++++ +title = "Distributed Convex Optimization for Electric Vehicle Aggregators" +year = 2017 +authors = ["José Rivera", "Christoph Goebel", "Hans-Arno Jacobsen"] +venue = "IEEE Transactions on Smart Grid" +publication_type = "Journal Article" +research = [] +external_url = "https://doi.org/10.1109/tsg.2015.2509030" +abstract = "One of the main challenges for electric vehicle (EV) aggregators is the definition of a control infrastructure that scales to large EV numbers. This paper proposes a new optimization framework for achieving computational scalability based on the alternating directions method of multipliers, which allows for distributing the optimization process across several servers/cores. We demonstrate the performance and versatility of our framework by applying it to two relevant aggregator objectives: 1) valley filling; and 2) cost-minimal charging with grid capacity constraints. Our results show that the solving time of our approach scales linearly with the number of controlled EVs and outperforms the centralized optimization benchmark as the fleet size becomes larger." ++++ diff --git a/content/publications/distributed-event-aggregation-for-content-based-publish-subscribe-systems.md b/content/publications/distributed-event-aggregation-for-content-based-publish-subscribe-systems.md new file mode 100644 index 0000000..e8a6e0d --- /dev/null +++ b/content/publications/distributed-event-aggregation-for-content-based-publish-subscribe-systems.md @@ -0,0 +1,10 @@ ++++ +title = "Distributed event aggregation for content-based publish/subscribe systems" +year = 2014 +authors = ["Navneet Kumar Pandey", "Kaiwen Zhang", "Stéphane Weiss", "Hans-Arno Jacobsen", "Roman Vitenberg"] +venue = "Proceedings of the 8th ACM International Conference on Distributed Event-Based Systems" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/2611286.2611302" +abstract = "Modern data-intensive applications handling massive event streams such as real-time traffic monitoring require support for both rich data filtering and aggregation. While the pub/sub communication paradigm provides an effective solution for the sought semantic diversity of event filtering, the event processing capabilities of existing pub/sub systems are restricted to singular event matching without support for stream aggregation, which so far can be accommodated only at the subscriber edge brokers." ++++ diff --git a/content/publications/distributed-quantum-computing-for-chemical-applications.md b/content/publications/distributed-quantum-computing-for-chemical-applications.md index f8254d4..648c632 100644 --- a/content/publications/distributed-quantum-computing-for-chemical-applications.md +++ b/content/publications/distributed-quantum-computing-for-chemical-applications.md @@ -2,16 +2,10 @@ title = "Distributed Quantum Computing for Chemical Applications" year = 2024 authors = ["Grier M. Jones", "Hans-Arno Jacobsen"] -venue = "arXiv Preprint" -publication_type = "ArXiv Preprint" +venue = "2024 IEEE International Conference on Quantum Computing and Engineering (QCE)" +publication_type = "Conference Paper" research = ["quantum-computing-systems"] tags = ["quantum-systems", "quantum-chemistry", "distributed-computing"] -summary = "Overview of distributed quantum computing with emphasis on chemistry-facing applications and current practical constraints." -external_url = "https://ieeexplore.ieee.org/abstract/document/10821041/" +external_url = "https://doi.org/10.1109/qce60285.2024.10270" +abstract = "In recent years, interest in quantum computing has increased due to technological advances in quantum hardware and algorithms. Despite the promises of quantum advantage, the applicability of quantum devices has been limited to few qubits on hardware that experiences decoherence due to noise. One proposed method to get around this challenge is distributed quantum computing (DQC). Like classical distributed computing, DQC aims at increasing compute power by spreading the compute processes across many devices, with the goal to minimize the noise and circuit depth required by quantum devices. In this paper, we cover the fundamental concepts of DQC and provide insight into where the field of DQC stands with respect to the field of chemistry -- a field which can potentially be used to demonstrate quantum advantage on noisy-intermediate scale quantum devices." +++ - -This publication frames distributed quantum computing as a way to work around -the limited size and noisy behavior of current quantum hardware. It surveys the -core ideas behind distributing quantum computation and considers how those ideas -map onto chemistry workloads, where the search for practical quantum advantage -is especially active. diff --git a/content/publications/distributed-ranked-data-dissemination-in-social-networks.md b/content/publications/distributed-ranked-data-dissemination-in-social-networks.md new file mode 100644 index 0000000..1922c08 --- /dev/null +++ b/content/publications/distributed-ranked-data-dissemination-in-social-networks.md @@ -0,0 +1,10 @@ ++++ +title = "Distributed Ranked Data Dissemination in Social Networks" +year = 2013 +authors = ["Kaiwen Zhang", "Mohammad Sadoghi", "Vinod Muthusamy", "Hans-Arno Jacobsen"] +venue = "2013 IEEE 33rd International Conference on Distributed Computing Systems" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icdcs.2013.19" +abstract = "The amount of content served on social networks can overwhelm users, who must sift through the data for relevant information. To facilitate users, we develop and implement dissemination of ranked data in social networks. Although top-k computation can be performed centrally at the user, the size of the event stream can constitute a significant bottleneck. Our approach distributes the top-k computation on an overlay network to reduce the number of events flowing through. Experiments performed using real Twitter and Facebook datasets with 5K and 30K query subscriptions demonstrate that social workloads exhibit properties that are advantageous for our solution." ++++ diff --git a/content/publications/distributed-stream-knn-join.md b/content/publications/distributed-stream-knn-join.md new file mode 100644 index 0000000..cce1cf3 --- /dev/null +++ b/content/publications/distributed-stream-knn-join.md @@ -0,0 +1,10 @@ ++++ +title = "Distributed Stream KNN Join" +year = 2021 +authors = ["Amirhesam Shahvarani", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 2021 International Conference on Management of Data" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/3448016.3457269" +abstract = "kNN join over data streams is an important operation for location-aware systems, which correlates events from different sources based on their occurrence locations. Combining the complexity of kNN join and the dynamicity of data streams, kNN join in streaming environments is a computationally intensive operator, and its performance can be greatly improved by utilizing the computational capabilities of modern non-uniform memory access (NUMA) computing platforms. However, the conventional approaches to kNN join for prestored datasets do not work efficiently with the kind of highly dynamic data found in streaming environments." ++++ diff --git a/content/publications/divide-and-conquer-algorithms-for-publish-subscribe-overlay-design.md b/content/publications/divide-and-conquer-algorithms-for-publish-subscribe-overlay-design.md new file mode 100644 index 0000000..209e111 --- /dev/null +++ b/content/publications/divide-and-conquer-algorithms-for-publish-subscribe-overlay-design.md @@ -0,0 +1,10 @@ ++++ +title = "Divide and Conquer Algorithms for Publish/Subscribe Overlay Design" +year = 2010 +authors = ["Chen Chen", "Hans-Arno Jacobsen", "Roman Vitenberg"] +venue = "2010 IEEE 30th International Conference on Distributed Computing Systems" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icdcs.2010.87" +abstract = "Overlay network design for topic-based publish/subscribe systems is of primary importance because the overlay directly impacts the system's performance. Determining a topic-connected overlay, in which for every topic the graph induced by nodes interested in the topic is connected, is a fundamental problem. Existing algorithms for this problem suffer from three key drawbacks: (1) prohibitively high running time cost, (2) requirement of full system knowledge and centralized operation, and (3) constructing overlay from scratch. From a practical point of view, these are all significant limitations. To address these concerns, in this paper, we develop novel algorithms that efficiently solve the problem of dynamically joining two or more topic-connected overlays. Inspired from the divide-and-conquer character of our approach, we derive an algorithm that solves the original problem at a fraction (up to 1.7%) of the running time cost of alternative solutions, but at the expense of an empirically insignificant increase in the average node degree." ++++ diff --git a/content/publications/dl-store-a-distributed-hybrid-oltp-and-olap-data-processing-engine.md b/content/publications/dl-store-a-distributed-hybrid-oltp-and-olap-data-processing-engine.md new file mode 100644 index 0000000..d13a3a8 --- /dev/null +++ b/content/publications/dl-store-a-distributed-hybrid-oltp-and-olap-data-processing-engine.md @@ -0,0 +1,10 @@ ++++ +title = "DL-Store: A Distributed Hybrid OLTP and OLAP Data Processing Engine" +year = 2016 +authors = ["Kaiwen Zhang", "Mohammad Sadoghi", "Hans-Arno Jacobsen"] +venue = "2016 IEEE 36th International Conference on Distributed Computing Systems (ICDCS)" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icdcs.2016.71" +abstract = "There has been a recent push in the database community towards supporting real-time analytical queries (OLAP) while sustaining a large volume of fine-grained updates (OLTP). Supporting these types of workloads require both an efficient data storage layer as well as a distributed architecture. In this demo, we address the latter point with our Distributed Lineage-based Data Store (DL-Store), which is a distributed data processing engine. DL-Store is built on top of L-Store, which is a lineage-based storage architecture designed to handle mixed OLTP and OLAP workloads, and provides scalability and elasticity by supporting multiple L-Store nodes. To maintain the desired consistency semantics, DL-Store employs a distributed transaction handler component which can horizontally scaled by provisioning additional transaction manager nodes. We leverage partitioning in the record space of the transactions to minimize communication across transaction managers while ensuring consistent execution. The demo shows our implementation of DL-Store over Apache Spark using a variety of use cases." ++++ diff --git a/content/publications/dualtable-a-hybrid-storage-model-for-update-optimization-in-hive.md b/content/publications/dualtable-a-hybrid-storage-model-for-update-optimization-in-hive.md new file mode 100644 index 0000000..53a08aa --- /dev/null +++ b/content/publications/dualtable-a-hybrid-storage-model-for-update-optimization-in-hive.md @@ -0,0 +1,10 @@ ++++ +title = "DualTable: A hybrid storage model for update optimization in Hive" +year = 2015 +authors = ["Songlin Hu", "Wantao Liu", "Tilmann Rabl", "Shuo Huang", "Ying Liang", "Zheng Xiao", "Hans-Arno Jacobsen", "Xubin Pei", "Jiye Wang"] +venue = "2015 IEEE 31st International Conference on Data Engineering" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icde.2015.7113381" +abstract = "Hive is the most mature and prevalent data warehouse tool providing SQL-like interface in the Hadoop ecosystem. It is successfully used in many Internet companies and shows its value for big data processing in traditional industries. However, enterprise big data processing systems as in Smart Grid applications usually require complicated business logics and involve many data manipulation operations like updates and deletes. Hive cannot offer sufficient support for these while preserving high query performance. Hive using the Hadoop Distributed File System (HDFS) for storage cannot implement data manipulation efficiently and Hive on HBase suffers from poor query performance even though it can support faster data manipulation. There is a project based on Hive issue Hive-5317 to support update operations, but it has not been finished in Hive's latest version. Since this ACID compliant extension adopts same data storage format on HDFS, the update performance problem is not solved. In this paper, we propose a hybrid storage model called DualTable, which combines the efficient streaming reads of HDFS and the random write capability of HBase. Hive on DualTable provides better data manipulation support and preserves query performance at the same time. Experiments on a TPC-H data set and on a real smart grid data set show that Hive on DualTable is up to 10 times faster than Hive when executing update and delete operations." ++++ diff --git a/content/publications/dynamic-load-balancing-in-distributed-content-based-publish-subscribe.md b/content/publications/dynamic-load-balancing-in-distributed-content-based-publish-subscribe.md new file mode 100644 index 0000000..d2bfe47 --- /dev/null +++ b/content/publications/dynamic-load-balancing-in-distributed-content-based-publish-subscribe.md @@ -0,0 +1,10 @@ ++++ +title = "Dynamic Load Balancing in Distributed Content-Based Publish/Subscribe" +year = 2006 +authors = ["Alex King Yeung Cheung", "Hans-Arno Jacobsen"] +venue = "Lecture Notes in Computer Science; Middleware 2006" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1007/11925071_8" +abstract = "Distributed content-based publish/subscribe systems to date suffer from performance degradation and poor scalability under uneven load distributions typical in real-world applications. The reason for this shortcoming is due to the lack of a load balancing solution, which have rarely been studied in the context of publish/subscribe. This paper proposes a load balancing solution specific to distributed content-based publish/subscribe systems that is distributed, dynamic, adaptive, transparent, and accommodates heterogeneity. The solution consists of three key contributions: a load balancing framework, a novel load estimation algorithm, and three offload strategies. Experimental results show that the proposed load balancing solution is efficient with less than 1.5 % overhead, effective with at least 91 % load estimation accuracy, and capable of distributing all of the system’s load originating from an edge point of the network. Keywords: Publish/subscribe, load distribution, content-based routing, load balancing, load estimation, subscriber migration, offloading algorithm 1" ++++ diff --git a/content/publications/dynamic-loss-based-sample-reweighting-for-improved-large-language-model-pretraining.md b/content/publications/dynamic-loss-based-sample-reweighting-for-improved-large-language-model-pretraining.md new file mode 100644 index 0000000..e3a072d --- /dev/null +++ b/content/publications/dynamic-loss-based-sample-reweighting-for-improved-large-language-model-pretraining.md @@ -0,0 +1,10 @@ ++++ +title = "Dynamic Loss-Based Sample Reweighting for Improved Large Language Model Pretraining" +year = 2025 +authors = ["Daouda Sow", "Herbert Woisetschläger", "Saikiran Bulusu", "Shiqiang Wang", "Hans-Arno Jacobsen", "Yingbin Liang"] +venue = "ICLR" +publication_type = "Conference Paper" +research = ["distributed-machine-learning"] +external_url = "https://openreview.net/forum?id=gU4ZgQNsOC" +abstract = "Pretraining large language models (LLMs) on vast and heterogeneous datasets is crucial for achieving state-of-the-art performance across diverse downstream tasks. However, current training paradigms treat all samples equally, overlooking the importance or relevance of individual samples throughout the training process. Existing reweighting strategies, which primarily focus on group-level data importance, fail to leverage fine-grained instance-level information and do not adapt dynamically to individual sample importance as training progresses. In this paper, we introduce novel algorithms for dynamic, instance-level data reweighting aimed at improving both the efficiency and effectiveness of LLM pretraining. Our methods adjust the weight of each training sample based on its loss value in an online fashion, allowing the model to dynamically focus on more informative or important samples at the current training stage. In particular, our framework allows us to systematically devise reweighting strategies deprioritizing redundant or uninformative data, which we find tend to work best. Furthermore, we develop a new theoretical framework for analyzing the impact of loss-based reweighting on the convergence of gradient-based optimization, providing the first formal characterization of how these strategies affect convergence bounds. We empirically validate our approach across a spectrum of tasks, from pretraining 7B and 1.4B parameter LLMs to smaller-scale language models and linear regression problems, demonstrating that our loss-based reweighting approach can lead to faster convergence and significantly improved performance." ++++ diff --git a/content/publications/effects-of-routing-computations-in-content-based-routing-networks-with-mobile-data-sources.md b/content/publications/effects-of-routing-computations-in-content-based-routing-networks-with-mobile-data-sources.md new file mode 100644 index 0000000..4c2b55a --- /dev/null +++ b/content/publications/effects-of-routing-computations-in-content-based-routing-networks-with-mobile-data-sources.md @@ -0,0 +1,10 @@ ++++ +title = "Effects of routing computations in content-based routing networks with mobile data sources" +year = 2005 +authors = ["Vinod Muthusamy", "Milenko Petrovic", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 11th annual international conference on Mobile computing and networking" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/1080829.1080840" +abstract = "This paper presents the first quantitative evaluation of the role of routing computations on performance when mobility is introduced to a content-based routing network. Additionally, the paper identifies the factors that affect the performance of a distributed publish/subscribe architecture supporting mobile publishers, formalizes publisher mobility protocols for distributed publish/subscribe systems, and develops and evaluates protocols that reduce the costs associated with supporting mobile publishers in publish/subscribe systems. Our results show that ignoring route computation time paints a false picture of the scalability of content-based routing networks, but that with appropriate protocols the adverse effects can be mitigated." ++++ diff --git a/content/publications/efficient-and-scalable-filtering-of-graph-based-metadata.md b/content/publications/efficient-and-scalable-filtering-of-graph-based-metadata.md new file mode 100644 index 0000000..d8b1557 --- /dev/null +++ b/content/publications/efficient-and-scalable-filtering-of-graph-based-metadata.md @@ -0,0 +1,9 @@ ++++ +title = "Efficient and scalable filtering of graph-based metadata" +year = 2005 +authors = ["Haifeng Liu", "Milenko Petrovic", "Hans-Arno Jacobsen"] +venue = "Journal of Web Semantics" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.1016/j.websem.2005.09.006" ++++ diff --git a/content/publications/efficient-constraint-processing-for-highly-personalized-location-based-services.md b/content/publications/efficient-constraint-processing-for-highly-personalized-location-based-services.md new file mode 100644 index 0000000..6108288 --- /dev/null +++ b/content/publications/efficient-constraint-processing-for-highly-personalized-location-based-services.md @@ -0,0 +1,9 @@ ++++ +title = "Efficient Constraint Processing for Highly Personalized Location Based Services" +year = 2004 +authors = ["Zhengdao Xu", "Hans-Arno Jacobsen"] +venue = "Proceedings 2004 VLDB Conference" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1016/b978-012088469-8.50128-5" ++++ diff --git a/content/publications/efficient-constraint-processing-for-location-aware-computing.md b/content/publications/efficient-constraint-processing-for-location-aware-computing.md new file mode 100644 index 0000000..a5254bf --- /dev/null +++ b/content/publications/efficient-constraint-processing-for-location-aware-computing.md @@ -0,0 +1,10 @@ ++++ +title = "Efficient constraint processing for location-aware computing" +year = 2005 +authors = ["Zhengdao Xu", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 6th international conference on Mobile data management" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/1071246.1071248" +abstract = "For many applications, such as friend finder, buddy tracking, and location mapping in mobile wireless networks or information sharing and cooperative caching in mobile ad hoc networks, it is often important to be able to identify whether a given set of moving objects is close to each other or close to a given point of demarcation. To achieve this, continuously available location position information of thousands of mobile objects must be correlated against each other to identify whether a fixed set of objects is in a certain proximity relation, which, if satisfied, would be signaled to the objects or any interested party. In this paper, we state this problem, referring to it as the location constraint matching problem and present and evaluate solutions for solving it. We introduce two types of location constraints to model the proximity relations and experimentally validate that our solution scales to the processing of hundreds of thousands of constraints and moving objects." ++++ diff --git a/content/publications/efficient-covering-for-top-k-filtering-in-content-based-publish-subscribe-systems.md b/content/publications/efficient-covering-for-top-k-filtering-in-content-based-publish-subscribe-systems.md new file mode 100644 index 0000000..4065842 --- /dev/null +++ b/content/publications/efficient-covering-for-top-k-filtering-in-content-based-publish-subscribe-systems.md @@ -0,0 +1,10 @@ ++++ +title = "Efficient covering for top-k filtering in content-based publish/subscribe systems" +year = 2017 +authors = ["Kaiwen Zhang", "Mohammad Sadoghi", "Vinod Muthusamy", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 18th ACM/IFIP/USENIX Middleware Conference" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/3135974.3135976" +abstract = "We investigate the use of content-based publish/subscribe for data dissemination in large-scale applications with expressive filtering requirements. In particular, we focus on top-k subscription filtering, where a publication is delivered only to the k best ranked subscribers, as ordered using expressive semantics such as relevance, fairness, and diversity. The naive approach to perform filtering early at the publisher edge works only if complete knowledge of the subscriptions is available, which is not compatible with the well-established covering optimization in scalable content-based publish/subscribe systems. We propose an efficient rank-cover technique to reconcile top-k subscription filtering with covering. We extend the covering model to support top-k and describe a novel algorithm for forwarding subscriptions to publishers while maintaining correctness. We also establish a framework for supporting different types of ranking semantics and propose an implementation to support fairness. Finally, we compare our solutions to a baseline covering system and perform sensitivity analysis to demonstrate that our optimized rank-cover algorithm retains both covering and fairness while achieving properties advantageous to our targeted workloads. In a typical setting, our optimized solution is scalable, selects fairly, and provides over 81% of the covering benefit." ++++ diff --git a/content/publications/efficient-data-transfer-in-shared-storage-cloud-data-processing-systems-with-optics.md b/content/publications/efficient-data-transfer-in-shared-storage-cloud-data-processing-systems-with-optics.md new file mode 100644 index 0000000..dfb8994 --- /dev/null +++ b/content/publications/efficient-data-transfer-in-shared-storage-cloud-data-processing-systems-with-optics.md @@ -0,0 +1,9 @@ ++++ +title = "Efficient Data Transfer in Shared-storage Cloud Data Processing Systems with OPTICS" +year = 2023 +authors = ["Gabriel Paulos", "Tongkun Zhang", "Yuqiu Zhang", "Gerry Zhu", "Jeyhun Karimov", "Hans-Arno Jacobsen"] +venue = "CASCON" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://dl.acm.org/doi/10.5555/3615924.3623630" ++++ diff --git a/content/publications/efficient-event-based-resource-discovery.md b/content/publications/efficient-event-based-resource-discovery.md index fd183f2..0bda7c4 100644 --- a/content/publications/efficient-event-based-resource-discovery.md +++ b/content/publications/efficient-event-based-resource-discovery.md @@ -1,19 +1,13 @@ +++ -title = "Efficient Event-Based Resource Discovery" +title = "Efficient event-based resource discovery" slug = "efficient-event-based-resource-discovery" year = 2009 -authors = ["Wei Yan", "Shuang Hu", "Vinod Muthusamy", "Hans-Arno Jacobsen"] -venue = "Proceedings of the 2009 ACM International Conference on Distributed Event-Based Systems (DEBS)" +authors = ["Wei Yan", "Songlin Hu", "Vinod Muthusamy", "Hans-Arno Jacobsen", "Li Zha"] +venue = "Proceedings of the Third ACM International Conference on Distributed Event-Based Systems" publication_type = "Conference Paper" research = ["data-management"] tags = ["resource-discovery", "publish-subscribe", "event-based-systems"] -summary = "Proposes a resource discovery framework based on the publish/subscribe model that supports both static and dynamic resources, with the companion workload of generated resource descriptions and subscription data." -external_url = "https://dl.acm.org/doi/abs/10.1145/1619258.1619284" +external_url = "https://doi.org/10.1145/1619258.1619284" related_datasets = ["resource-discovery-workload"] +abstract = "The ability to find services or resources that satisfy some criteria is an important aspect of distributed systems. This paper presents an event-based architecture to support more dynamic discovery scenarios, including efficient discovery of resources whose attributes can change, and continuous monitoring for resources that satisfy a set of constraints. Furthermore, algorithms are developed to optimize the discovery cost by reusing results among similar concurrent discovery requests. Detailed evaluations under various workload distributions demonstrate the feasibility of the architecture and show significant benefits of the optimizations in terms of network traffic and discovery processing time." +++ - -Resource discovery is a recurring building block in large-scale distributed systems. -This paper proposes a resource discovery framework based on the publish/subscribe -model that supports both static and dynamic resources. The companion workload -package contains generated resource descriptions and subscription data used in the -experiments. diff --git a/content/publications/efficient-event-processing-through-reconfigurable-hardware-for-algorithmic-trading.md b/content/publications/efficient-event-processing-through-reconfigurable-hardware-for-algorithmic-trading.md new file mode 100644 index 0000000..c3ea005 --- /dev/null +++ b/content/publications/efficient-event-processing-through-reconfigurable-hardware-for-algorithmic-trading.md @@ -0,0 +1,10 @@ ++++ +title = "Efficient Event Processing through Reconfigurable Hardware for Algorithmic Trading" +year = 2010 +authors = ["Mohammad Sadoghi", "Hans-Arno Jacobsen", "Martin Labrecque", "Warren Shum", "Harsh Singh"] +venue = "Proceedings of the VLDB Endowment" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.14778/1920841.1921029" +abstract = "In this demo, we present fpga-ToPSS (Toronto Publish/Subscribe System Family), an efficient event processing platform for high-frequency and low-latency algorithmic trading. Our event processing platform is built over reconfigurable hardware---FPGAs---to achieve line-rate processing. Furthermore, our event processing engine supports Boolean expression matching with an expressive predicate language that models complex financial strategies to autonomously buy and sell stocks based on real-time financial data." ++++ diff --git a/content/publications/efficient-federated-learning-methods-for-foundation-model-training.md b/content/publications/efficient-federated-learning-methods-for-foundation-model-training.md index d5b8de4..7859fea 100644 --- a/content/publications/efficient-federated-learning-methods-for-foundation-model-training.md +++ b/content/publications/efficient-federated-learning-methods-for-foundation-model-training.md @@ -1,17 +1,11 @@ +++ title = "A Survey on Efficient Federated Learning Methods for Foundation Model Training" year = 2024 -authors = ["Herbert Woisetschlaeger", "Alexander Erben", "Shiqiang Wang", "Ruben Mayer", "Hans-Arno Jacobsen"] -venue = "Proceedings of the Thirty-Third International Joint Conference on Artificial Intelligence" +authors = ["Herbert Woisetschläger", "Alexander Erben", "Shiqiang Wang", "Ruben Mayer", "Hans-Arno Jacobsen"] +venue = "Proceedings of the Thirty-Third International Joint Conference on Artificial Intelligence (IJCAI)" publication_type = "Conference Paper" research = ["distributed-machine-learning"] tags = ["federated-learning", "foundation-models", "systems-survey"] -summary = "Survey of computationally and communication-efficient federated learning strategies for foundation-model settings." -external_url = "https://arxiv.org/abs/2401.04472" +external_url = "https://www.ijcai.org/proceedings/2024/919" +abstract = "Federated Learning (FL) has become an established technique to facilitate privacy-preserving collaborative training across a multitude of clients. However, new approaches to FL often discuss their contributions involving small deep-learning models only and focus on training full models on clients. In the wake of Foundation Models (FM), the reality is different for many deep learning applications. Typically, FMs have already been pre-trained across a wide variety of tasks and can be fine-tuned to specific downstream tasks over significantly smaller datasets than required for full model training. However, access to such datasets is often challenging. By its design, FL can help to open data silos. With this survey, we introduce a novel taxonomy focused on computational and communication efficiency, the vital elements to make use of FMs in FL systems. We discuss the benefits and drawbacks of parameter-efficient fine-tuning (PEFT) for FL applications, elaborate on the readiness of FL frameworks to work with FMs, and provide future research opportunities on how to evaluate generative models in FL as well as the interplay of privacy and PEFT." +++ - -This survey looks at what changes when federated learning meets foundation -models. Rather than assuming small models and full-model training on each -client, it focuses on efficiency questions that become central once pre-trained -models, parameter-efficient fine-tuning, and communication costs dominate the -design space. diff --git a/content/publications/efficiently-mining-crosscutting-concerns-through-random-walks.md b/content/publications/efficiently-mining-crosscutting-concerns-through-random-walks.md new file mode 100644 index 0000000..974fffa --- /dev/null +++ b/content/publications/efficiently-mining-crosscutting-concerns-through-random-walks.md @@ -0,0 +1,10 @@ ++++ +title = "Efficiently mining crosscutting concerns through random walks" +year = 2007 +authors = ["Charles Zhang", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 6th international conference on Aspect-oriented software development" +publication_type = "Conference Paper" +research = [] +external_url = "https://doi.org/10.1145/1218563.1218588" +abstract = "Inspired by our past manual aspect mining experiences, this paper describes a random walk model to approximate how crosscutting concerns can be discovered in the absence of domain knowledge of the investigated application. Random walks are performed on the coupling graphs extracted from the program sources. The ideas underlying the popular page-rank algorithm are adapted and extended to generate ranks reflecting the degrees of \"popularity\" and \"significance\" for each of the program elements on the coupling graphs. Filtering techniques, exploiting both types of ranks, are applied to produce a final list of candidates representing crosscutting concerns. The resulting aspect mining algorithm is evaluated on numerous Java applications ranging from a small-scale drawing application, to a medium-sized middleware application, and to a largescale enterprise application server. In seconds, the aspect mining algorithm is able to produce results comparable to our prior manual mining efforts. The mining algorithm also proves effective in helping domain experts identify latent crosscutting concerns." ++++ diff --git a/content/publications/electrical-appliance-classification-using-deep-convolutional-neural-networks-on-high-frequency-current-measurements.md b/content/publications/electrical-appliance-classification-using-deep-convolutional-neural-networks-on-high-frequency-current-measurements.md new file mode 100644 index 0000000..32b1ef7 --- /dev/null +++ b/content/publications/electrical-appliance-classification-using-deep-convolutional-neural-networks-on-high-frequency-current-measurements.md @@ -0,0 +1,10 @@ ++++ +title = "Electrical Appliance Classification using Deep Convolutional Neural Networks on High Frequency Current Measurements" +year = 2018 +authors = ["Daniel Jorde", "Thomas Kriechbaumer", "Hans-Arno Jacobsen"] +venue = "2018 IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids (SmartGridComm)" +publication_type = "Conference Paper" +research = ["distributed-machine-learning"] +external_url = "https://doi.org/10.1109/smartgridcomm.2018.8587452" +abstract = "Monitoring the energy demand of appliances can raise consumer awareness and therefore reduce energy consumption. Using a single-point measurement of mains energy consumption can keep costs and hardware complexity to a minimum. This data stream of raw voltage and current measurements can be used in machine learning tasks to extract information. We apply Deep Convolutional Neural Networks on an electrical appliance classification task, using raw high frequency start up events from two datasets. We further present Data Augmentation techniques to improve the model performance and evaluate different data normalization techniques. We achieve a perfect classification on WHITED and a Fl-Score of 0.69 on PLAID." ++++ diff --git a/content/publications/embedpart-embedding-driven-graph-partitioning-for-scalable-graph-neural-network-training.md b/content/publications/embedpart-embedding-driven-graph-partitioning-for-scalable-graph-neural-network-training.md new file mode 100644 index 0000000..e786ce3 --- /dev/null +++ b/content/publications/embedpart-embedding-driven-graph-partitioning-for-scalable-graph-neural-network-training.md @@ -0,0 +1,10 @@ ++++ +title = "EmbedPart: Embedding-Driven Graph Partitioning for Scalable Graph Neural Network Training" +year = 2026 +authors = ["Nikolai Merkel", "Ruben Mayer", "Volker Markl", "Hans-Arno Jacobsen"] +venue = "arXiv" +publication_type = "ArXiv Preprint" +research = ["distributed-machine-learning", "data-management"] +external_url = "https://arxiv.org/abs/2604.01000" +abstract = "Graph Neural Networks (GNNs) are widely used for learning on graph-structured data, but scaling GNN training to massive graphs remains challenging. To enable scalable distributed training, graphs are divided into smaller partitions that are distributed across multiple machines such that inter-machine communication is minimized and computational load is balanced. In practice, existing partitioning approaches face a fundamental trade-off between partitioning overhead and partitioning quality. We propose EmbedPart, an embedding-driven partitioning approach that achieves both speed and quality. Instead of operating directly on irregular graph structures, EmbedPart leverages node embeddings produced during the actual GNN training workload and clusters these dense embeddings to derive a partitioning. EmbedPart achieves more than 100x speedup over Metis while maintaining competitive partitioning quality and accelerating distributed GNN training. Moreover, EmbedPart naturally supports graph updates and fast repartitioning, and can be applied to graph reordering to improve data locality and accelerate single-machine GNN training. By shifting partitioning from irregular graph structures to dense embeddings, EmbedPart enables scalable and high-quality graph data optimization." ++++ diff --git a/content/publications/enclavecache-a-secure-and-scalable-key-value-cache-in-multi-tenant-clouds-using-intel-sgx.md b/content/publications/enclavecache-a-secure-and-scalable-key-value-cache-in-multi-tenant-clouds-using-intel-sgx.md new file mode 100644 index 0000000..ddf1707 --- /dev/null +++ b/content/publications/enclavecache-a-secure-and-scalable-key-value-cache-in-multi-tenant-clouds-using-intel-sgx.md @@ -0,0 +1,10 @@ ++++ +title = "EnclaveCache: A Secure and Scalable Key-value Cache in Multi-tenant Clouds using Intel SGX" +year = 2019 +authors = ["Li-Xia Chen", "Jian Li", "Ruhui Ma", "Haibing Guan", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 20th International Middleware Conference" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/3361525.3361533" +abstract = "With in-memory key-value caches such as Redis and Memcached being a key component for many systems to improve throughput and reduce latency, cloud caches have been widely adopted for small companies to deploy their own cache systems. However, data security is still a major concern, which affects the adoption of cloud caches. Tenant's data stored in a multi-tenant cloud environment faces threats from both co-located other tenants, as well as the untrusted cloud provider." ++++ diff --git a/content/publications/energieinformatik-aktuelle-und-zukunftige-forschungsschwerpunkte.md b/content/publications/energieinformatik-aktuelle-und-zukunftige-forschungsschwerpunkte.md new file mode 100644 index 0000000..7c78218 --- /dev/null +++ b/content/publications/energieinformatik-aktuelle-und-zukunftige-forschungsschwerpunkte.md @@ -0,0 +1,9 @@ ++++ +title = "Energieinformatik - Aktuelle und zukünftige Forschungsschwerpunkte" +year = 2014 +authors = ["Christoph Goebel", "Hans-Arno Jacobsen", "Victor del Razo", "Christoph Doblander", "José Rivera", "Jens P. Ilg", "Christoph M. Flath", "Hartmut Schmeck", "Christof Weinhardt", "Daniel Pathmaperuma", "Hans-Jürgen Appelrath", "Michael Sonnenschein", "Sebastian Lehnhoff", "Oliver Kramer", "Thorsten Staake", "Elgar Fleisch", "Dirk Neumann", "Jens Strüker", "Koray Erek", "Rüdiger Zarnekow", "Holger Ziekow", "Jörg Lässig"] +venue = "WIRTSCHAFTSINFORMATIK" +publication_type = "Journal Article" +research = [] +external_url = "https://doi.org/10.1007/s11576-013-0396-9" ++++ diff --git a/content/publications/energy-informatics-current-and-future-research-directions.md b/content/publications/energy-informatics-current-and-future-research-directions.md new file mode 100644 index 0000000..b7fc525 --- /dev/null +++ b/content/publications/energy-informatics-current-and-future-research-directions.md @@ -0,0 +1,9 @@ ++++ +title = "Energy Informatics - Current and Future Research Directions" +year = 2014 +authors = ["Christoph Goebel", "Hans-Arno Jacobsen", "Victor del Razo", "Christoph Doblander", "José Rivera", "Jens P. Ilg", "Christoph M. Flath", "Hartmut Schmeck", "Christof Weinhardt", "Daniel Pathmaperuma", "Hans-Jürgen Appelrath", "Michael Sonnenschein", "Sebastian Lehnhoff", "Oliver Kramer", "Thorsten Staake", "Elgar Fleisch", "Dirk Neumann", "Jens Strüker", "Koray Erek", "Rüdiger Zarnekow", "Holger Ziekow", "Jörg Lässig"] +venue = "Business & Information Systems Engineering" +publication_type = "Journal Article" +research = [] +external_url = "https://doi.org/10.1007/s12599-013-0304-2" ++++ diff --git a/content/publications/energy-vs-privacy-estimating-the-ecological-impact-of-federated-learning.md b/content/publications/energy-vs-privacy-estimating-the-ecological-impact-of-federated-learning.md new file mode 100644 index 0000000..5d90d5d --- /dev/null +++ b/content/publications/energy-vs-privacy-estimating-the-ecological-impact-of-federated-learning.md @@ -0,0 +1,10 @@ ++++ +title = "Energy vs Privacy: Estimating the Ecological Impact of Federated Learning" +year = 2023 +authors = ["René Schwermer", "Ruben Mayer", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 14th ACM International Conference on Future Energy Systems" +publication_type = "Conference Paper" +research = ["distributed-machine-learning"] +external_url = "https://doi.org/10.1145/3575813.3597344" +abstract = "The increasing usage of edge devices and stricter data privacy regulations motivate the use of federated learning (FL). At the same time, more and more stakeholders are concerned about the ecological impact of machine learning and its associate network traffic. The current research in FL does not investigate the impact of different network constraints and privacy-enhancing techniques, such as differential privacy, on the network traffic and energy consumption of the clients. Most experiments run either on virtual machines or on one machine with simulated clients. In such environments, it is challenging to measure each client’s network and energy usage. Therefore, we built our \"Distributed Edge Device Testbed\" (DEDT) and evaluate a convolutional neural network trained on the MNIST data set under different network constraints on DEDT, with differential privacy and with an increasing amount of participating clients. For each experiment, we quantify the network traffic, energy consumption, and training time. The results show the importance of experiments on physically separated nodes and the need to improve software-based power monitoring. The estimated energy consumption deviates by up to 35 % from the measured ones. The accuracy of the estimated network traffic depends on the monitored network interface and gives an error of 18 % for virtual machines in combination with monitoring the Ethernet interface. The training time also increases linearly with the number of participating clients." ++++ diff --git a/content/publications/environmental-footprints-of-query-processing-a-vision-for-sustainable-database-architectures.md b/content/publications/environmental-footprints-of-query-processing-a-vision-for-sustainable-database-architectures.md new file mode 100644 index 0000000..5d455dd --- /dev/null +++ b/content/publications/environmental-footprints-of-query-processing-a-vision-for-sustainable-database-architectures.md @@ -0,0 +1,10 @@ ++++ +title = "Environmental Footprints of Query Processing: A Vision for Sustainable Database Architectures" +year = 2025 +authors = ["Michail Bachras", "Hans-Arno Jacobsen"] +venue = "Proceedings of the VLDB Endowment" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.14778/3749646.3749676" +abstract = "Database systems underpin modern computing infrastructure, yet their environmental impact remains a significant blind spot in both industry and research. As data volumes grow exponentially, the energy consumption, carbon emissions, and water usage of database operations increasingly threaten global sustainability goals. Our paper explores this multidimensional environmental footprint and proposes a vision where sustainability becomes a first-class design criterion alongside traditional performance metrics. We reimagine database architectures that incorporate environmental awareness throughout both hardware and software layers. By identifying critical research challenges, we establish a foundation for database systems that can deliver high performance while meeting the environmental demands of our resource-constrained world." ++++ diff --git a/content/publications/epoch-based-optimistic-concurrency-control-in-geo-replicated-databases.md b/content/publications/epoch-based-optimistic-concurrency-control-in-geo-replicated-databases.md new file mode 100644 index 0000000..de2748f --- /dev/null +++ b/content/publications/epoch-based-optimistic-concurrency-control-in-geo-replicated-databases.md @@ -0,0 +1,10 @@ ++++ +title = "Epoch-based Optimistic Concurrency Control in Geo-replicated Databases" +year = 2026 +authors = ["Yunhao Mao", "Harunari Takata", "Michail Bachras", "Yuqiu Zhang", "Shiquan Zhang", "Gengrui Zhang", "Hans-Arno Jacobsen"] +venue = "Proceedings of the ACM on Management of Data" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.1145/3802052" +abstract = "Achieving high-performance transaction processing in geo-replicated OLTP databases is challenging due to the extensive over-coordination in distributed atomic commitment, concurrency control, and fault-tolerant replication protocols. To address this issue, we introduce Minerva, a unified distributed concurrency control protocol designed for highly scalable multi-leader replication. Minerva employs a novel epoch-based asynchronous replication protocol that decouples data propagation from the commitment process, enabling continuous transaction replication. Optimistic concurrency control is used to allow replicas to execute transactions concurrently and to commit without coordination. For conflict detection during validation, we construct a conflict graph and use a maximum weight independent set search algorithm to select the optimal subset of non-conflicting transactions for commitment, minimizing the number of invalid transactions. Finally, we deterministically re-execute conflicting transactions, ensuring serializability while eliminating aborts. Our evaluation demonstrates that Minerva outperforms state-of-the-art replicated databases, achieving over 3x higher throughput in scalability experiments and 2.8x higher throughput in a high-latency network simulation with the TPC-C benchmark." ++++ diff --git a/content/publications/eqosystem-supporting-fluid-distributed-service-oriented-workflows.md b/content/publications/eqosystem-supporting-fluid-distributed-service-oriented-workflows.md new file mode 100644 index 0000000..19ab64b --- /dev/null +++ b/content/publications/eqosystem-supporting-fluid-distributed-service-oriented-workflows.md @@ -0,0 +1,9 @@ ++++ +title = "eQoSystem: supporting fluid distributed service-oriented workflows" +year = 2011 +authors = ["Vinod Muthusamy", "Young Yoon", "Mohammad Sadoghi", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 5th ACM international conference on Distributed event-based system" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/2002259.2002320" ++++ diff --git a/content/publications/escape-to-precaution-against-leader-failures.md b/content/publications/escape-to-precaution-against-leader-failures.md new file mode 100644 index 0000000..62be767 --- /dev/null +++ b/content/publications/escape-to-precaution-against-leader-failures.md @@ -0,0 +1,10 @@ ++++ +title = "ESCAPE to Precaution against Leader Failures" +year = 2022 +authors = ["Gengrui Zhang", "Hans-Arno Jacobsen"] +venue = "2022 IEEE 42nd International Conference on Distributed Computing Systems (ICDCS)" +publication_type = "Conference Paper" +research = [] +external_url = "https://doi.org/10.1109/icdcs54860.2022.00066" +abstract = "Leader-based consensus protocols must undergo a view-change phase to elect a new leader when the current leader fails. The new leader often comes from a candidate server that collects votes from a quorum of servers. However, voting-based election mechanisms intrinsically incite competition in leadership candidacy since candidates may collect only partial votes. This split-vote scenario can result in no leadership winner and thus prolongs the undesired view-change period. In this paper, we investigate a case study of Raft’s leader election and propose a new leader election protocol, called ESCAPE, that fundamentally solves split votes by prioritizing servers based on their log responsiveness. ESCAPE dynamically distributes configurations that offer different priorities to servers through periodic heartbeats. In each assignment, ESCAPE assigns configurations that are more inclined to win an election to servers that have more up-to-date log responsiveness, thereby preparing a pool of prioritized candidates. Consequently, when the next election takes place, the candidate with the highest priority can defeat its counterparts and becomes the next leader without competition. The evaluation results show that ESCAPE progressively reduces the leader election time when the cluster scales up, and the improvement becomes more significant under message loss." ++++ diff --git a/content/publications/eva-fair-and-auditable-electric-vehicle-charging-service-using-blockchain.md b/content/publications/eva-fair-and-auditable-electric-vehicle-charging-service-using-blockchain.md new file mode 100644 index 0000000..9ea1f18 --- /dev/null +++ b/content/publications/eva-fair-and-auditable-electric-vehicle-charging-service-using-blockchain.md @@ -0,0 +1,10 @@ ++++ +title = "EVA: Fair and Auditable Electric Vehicle Charging Service using Blockchain" +year = 2018 +authors = ["Jelena Pajic", "José Rivera", "Kaiwen Zhang", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 12th ACM International Conference on Distributed and Event-based Systems" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/3210284.3219776" +abstract = "The recent success of electric vehicles leads to unprecedentedly high peaks of demand on the electric grid at the times when most people charge their cars. In order to avoid unreasonably rising costs due to inefficient utilization of the electricity infrastructure, we propose EVA: a scheduling system to solve the valley filling problem by distributing the electricity demand generated by electric vehicles in a geographically limited area efficiently over time spans in which the electric grid is underutilized. EVA is based on a smart contract running on the Ethereum blockchain in combination with off-chain computational nodes performing the schedule calculation using the Alternating Direction Method of Multipliers (ADMM). This allows for a high degree of transparency and verifiability in the scheduling computation results while maintaining a reasonable level of efficiency. In order to interact with the scheduling system, we developed a decentralized app with a graphical frontend, where the user can enter vehicle information and future energy requirements as well as review upcoming schedules. The calculation of the schedule is performed on a daily basis, continuously providing schedules for participating users for the following day." ++++ diff --git a/content/publications/evaluating-interwell-connectivity-in-waterflooding-reservoirs-with-graph-based-cooperation-mission-neural-networks.md b/content/publications/evaluating-interwell-connectivity-in-waterflooding-reservoirs-with-graph-based-cooperation-mission-neural-networks.md new file mode 100644 index 0000000..93abf95 --- /dev/null +++ b/content/publications/evaluating-interwell-connectivity-in-waterflooding-reservoirs-with-graph-based-cooperation-mission-neural-networks.md @@ -0,0 +1,10 @@ ++++ +title = "Evaluating Interwell Connectivity in Waterflooding Reservoirs with Graph-Based Cooperation-Mission Neural Networks" +year = 2022 +authors = ["Xingjie Zeng", "Weishan Zhang", "Tao Chen", "Hans-Arno Jacobsen", "Jiehan Zhou", "Bingyang Chen"] +venue = "SPE Journal" +publication_type = "Journal Article" +research = ["distributed-machine-learning"] +external_url = "https://doi.org/10.2118/209607-pa" +abstract = "Summary Interwell connectivity plays a key role in waterflooding for guiding water injection. The existing works focus on the response relationship between one injection well and one production well. No research has explored the structural information of waterflooding on a well pattern. To address this challenge, this paper proposes cooperation-mission neural networks for interwell connectivity with graph information. Specifically, we propose some assumptions based on the petroleum domain to represent the well pattern with an adjacent matrix of the graph. Then we propose two targets from the view of injection well groups and production well groups. Accordingly, we propose cooperation-mission neural networks from these two aspects to evaluate the interwell connectivity in the well pattern. We test our model from two perspectives: the accuracy of estimation with tracer and the graduality of interwell connectivity. The results demonstrate that our model makes a good performance and achieves the connectivity analysis accuracy rate of 91.4%. Moreover, this study demonstrates that it is practical to evaluate the interwell connectivity with graph." ++++ diff --git a/content/publications/evaluating-proximity-relations-under-uncertainty.md b/content/publications/evaluating-proximity-relations-under-uncertainty.md new file mode 100644 index 0000000..c96740a --- /dev/null +++ b/content/publications/evaluating-proximity-relations-under-uncertainty.md @@ -0,0 +1,10 @@ ++++ +title = "Evaluating Proximity Relations Under Uncertainty" +year = 2007 +authors = ["Zhengdao Xu", "Hans-Arno Jacobsen"] +venue = "2007 IEEE 23rd International Conference on Data Engineering" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icde.2007.367933" +abstract = "For location-based services it is often essential to efficiently process proximity relations among mobile objects, such as to establish whether a group of friends or family members are within a given distance of each other A severe limitation in accurately establishing such relations is the inaccuracy of dynamically obtained position data, the point in time, and the frequency with which the position data is collected. In this paper, we use the common model of interpreting the unknown position of an object by a probability distribution centered around the last know position of the object. While this approach is straight forward, it poses severe difficulties for establishing the truth or falsehood of the proximity relation. To address this problem, we analytically quantify the lower and upper bounds of the size of the smallest circle that covers the mobile objects involved in the proximity relation. Based on this result we propose two novel algorithms that closely monitor the relation at low location update cost. Furthermore, we develop a cost-effective estimation technique to determine the probability of match for a given proximity relation." ++++ diff --git a/content/publications/event-detection-for-energy-consumption-monitoring.md b/content/publications/event-detection-for-energy-consumption-monitoring.md new file mode 100644 index 0000000..76731a4 --- /dev/null +++ b/content/publications/event-detection-for-energy-consumption-monitoring.md @@ -0,0 +1,10 @@ ++++ +title = "Event Detection for Energy Consumption Monitoring" +year = 2021 +authors = ["Daniel Jorde", "Hans-Arno Jacobsen"] +venue = "IEEE Transactions on Sustainable Computing" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.1109/tsusc.2020.3012066" +abstract = "The accurate detection of appliance state transitions in electrical signals is fundamental for numerous energy-conserving applications. We present an extensive overview and categorization of the current state in event detection on high-sampling-rate signals. Existing approaches are designed for specific environments and need to be tediously adapted for new ones. Thus, we propose an unsupervised, multi-environment event detector, outperforming four state-of-the-art algorithms on two heterogeneous public datasets." ++++ diff --git a/content/publications/event-exposure-for-web-services-a-grey-box-approach-to-compose-and-evolve-web-services.md b/content/publications/event-exposure-for-web-services-a-grey-box-approach-to-compose-and-evolve-web-services.md new file mode 100644 index 0000000..7f2221c --- /dev/null +++ b/content/publications/event-exposure-for-web-services-a-grey-box-approach-to-compose-and-evolve-web-services.md @@ -0,0 +1,9 @@ ++++ +title = "Event Exposure for Web Services: A Grey-Box Approach to Compose and Evolve Web Services" +year = 2010 +authors = ["Chunyang Ye", "Hans-Arno Jacobsen"] +venue = "Lecture Notes in Computer Science; The Smart Internet" +publication_type = "Book Chapter" +research = ["data-management"] +external_url = "https://doi.org/10.1007/978-3-642-16599-3_14" ++++ diff --git a/content/publications/evolving-pub-sub-subscriptions-for-multiplayer-online-games-demo.md b/content/publications/evolving-pub-sub-subscriptions-for-multiplayer-online-games-demo.md new file mode 100644 index 0000000..90520c5 --- /dev/null +++ b/content/publications/evolving-pub-sub-subscriptions-for-multiplayer-online-games-demo.md @@ -0,0 +1,10 @@ ++++ +title = "Evolving pub/sub subscriptions for multiplayer online games: demo" +year = 2016 +authors = ["César Cañas", "Kaiwen Zhang", "Bettina Kemme", "Jörg Kienzle", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 10th ACM International Conference on Distributed and Event-based Systems" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/2933267.2933297" +abstract = "We identify a class of content-based pub/sub applications with highly dynamic subscriptions. This includes location-based notification systems, predictive stock trading, and multiplayer games. The traditional method of handling subscription changes by engaging in expensive re-subscription protocols is inadequate when the workload is sufficiently large." ++++ diff --git a/content/publications/experiment-benchmark-paper-to-what-extent-does-quality-matter-the-impact-of-graph-data-quality-on-gnn-model-performance.md b/content/publications/experiment-benchmark-paper-to-what-extent-does-quality-matter-the-impact-of-graph-data-quality-on-gnn-model-performance.md new file mode 100644 index 0000000..e7fd485 --- /dev/null +++ b/content/publications/experiment-benchmark-paper-to-what-extent-does-quality-matter-the-impact-of-graph-data-quality-on-gnn-model-performance.md @@ -0,0 +1,11 @@ ++++ +title = "Experiment & Benchmark Paper: To What Extent Does Quality Matter? The Impact of Graph Data Quality on GNN Model Performance" +year = 2025 +authors = ["Jana Vatter", "Maurice L. Rochau", "Ruben Mayer", "Hans-Arno Jacobsen"] +venue = "VLDB Workshops" +publication_type = "Conference Paper" +research = ["distributed-machine-learning"] +external_url = "https://www.vldb.org/2025/Workshops/VLDB-Workshops-2025/LSGDA/LSGDA25_05.pdf" +abstract = "Real-world data often is noisy and error-prone which can negatively influence machine learning models. Graph Neural Networks (GNNs) introduce the additional challenge that during training, node information is iteratively passed through the graph along the edges. Consequently, errors or deviations in the graph data could highly impact the model’s predictive capability. Our work systematically investigates how quality deviations in graph datasets influence the GNN model performance. We focus on the node features and explore three dimensions: the rate of modified features, the amplitude of modification, and the feature precision. Based on our results, we give insights and recommendations for practitioners. For instance, when using highly clustered graphs, modifying around 40% of the features only results in a slight decrease of performance and the rate of modified features is more crucial than the amplitude of modification. We illustrate practical implications of our results by establishing connections to real world domains such as graph dataset acquisition and efficient GNN training." +abstract_license_url = "https://creativecommons.org/licenses/by-nc-nd/4.0/" ++++ diff --git a/content/publications/expressive-location-based-continuous-query-evaluation-with-binary-decision-diagrams.md b/content/publications/expressive-location-based-continuous-query-evaluation-with-binary-decision-diagrams.md new file mode 100644 index 0000000..57b9f88 --- /dev/null +++ b/content/publications/expressive-location-based-continuous-query-evaluation-with-binary-decision-diagrams.md @@ -0,0 +1,10 @@ ++++ +title = "Expressive Location-Based Continuous Query Evaluation with Binary Decision Diagrams" +year = 2009 +authors = ["Zhengdao Xu", "Hans-Arno Jacobsen"] +venue = "2009 IEEE 25th International Conference on Data Engineering" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icde.2009.189" +abstract = "This paper is concerned with developing algorithms to enable the use of a rich query language supporting spatio-temporal processing among moving objects. The queries we consider monitor location constraints. A location constraint represents a proximity relation among moving objects and among moving and static objects. Location constraint processing is like continuous query processing; once the location constraint is submitted to the system, it remains active until explicitly revoked. Location updates that represent the movement of objects are streamed into the system and trigger the evaluation of all location constraints stored with the system. Matching constraints are communicated back to interested subscribers." ++++ diff --git a/content/publications/externalizing-java-server-concurrency-with-cal.md b/content/publications/externalizing-java-server-concurrency-with-cal.md new file mode 100644 index 0000000..267c831 --- /dev/null +++ b/content/publications/externalizing-java-server-concurrency-with-cal.md @@ -0,0 +1,9 @@ ++++ +title = "Externalizing Java Server Concurrency with CAL" +year = 2008 +authors = ["Charles Zhang", "Hans-Arno Jacobsen"] +venue = "Lecture Notes in Computer Science; ECOOP 2008 – Object-Oriented Programming" +publication_type = "Conference Paper" +research = [] +external_url = "https://doi.org/10.1007/978-3-540-70592-5_16" ++++ diff --git a/content/publications/fabriccrdt-a-conflict-free-replicated-datatypes-approach-to-permissioned-blockchains.md b/content/publications/fabriccrdt-a-conflict-free-replicated-datatypes-approach-to-permissioned-blockchains.md new file mode 100644 index 0000000..655432c --- /dev/null +++ b/content/publications/fabriccrdt-a-conflict-free-replicated-datatypes-approach-to-permissioned-blockchains.md @@ -0,0 +1,10 @@ ++++ +title = "FabricCRDT: A Conflict-Free Replicated Datatypes Approach to Permissioned Blockchains" +year = 2019 +authors = ["Pezhman Nasirifard", "Ruben Mayer", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 20th International Middleware Conference" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/3361525.3361540" +abstract = "With the increased adaption of blockchain technologies, permissioned blockchains such as Hyperledger Fabric provide a robust ecosystem for developing production-grade decentralized applications. However, the additional latency between executing and committing transactions, due to Fabric's three-phase transaction lifecycle of Execute-Order-Validate (EOV), is a potential scalability bottleneck. The added latency increases the probability of concurrent updates on the same keys by different transactions, leading to transaction failures caused by Fabric's concurrency control mechanism. The transaction failures increase the application development complexity and decrease Fabric's throughput. Conflict-free Replicated Datatypes (CRDTs) provide a solution for merging and resolving conflicts in the presence of concurrent updates. In this work, we introduce FabricCRDT, an approach for integrating CRDTs to Fabric. Our evaluations show that in general, FabricCRDT offers higher throughput of successful transactions than Fabric, while successfully committing and merging all conflicting transactions without any failures." ++++ diff --git a/content/publications/fabricunit-a-framework-for-faster-execution-of-unit-tests-on-hyperledger-fabric.md b/content/publications/fabricunit-a-framework-for-faster-execution-of-unit-tests-on-hyperledger-fabric.md index 77e8f07..68c2578 100644 --- a/content/publications/fabricunit-a-framework-for-faster-execution-of-unit-tests-on-hyperledger-fabric.md +++ b/content/publications/fabricunit-a-framework-for-faster-execution-of-unit-tests-on-hyperledger-fabric.md @@ -1,11 +1,11 @@ +++ -title = "Fabricunit: A framework for faster execution of unit tests on hyperledger fabric" +title = "FabricUnit: A Framework for Faster Execution of Unit Tests on Hyperledger Fabric" year = 2020 -authors = ["Shashank Motepalli", "Patricia Vilain", "Hans-Arno Jacobsen"] +authors = ["Shashank Motepalli", "Patrícia Vilain", "Hans-Arno Jacobsen"] venue = "2020 IEEE International Conference on Blockchain and Cryptocurrency (ICBC)" publication_type = "Conference Paper" research = ["data-management"] tags = ["blockchain"] -external_url = "https://ieeexplore.ieee.org/document/9169430/" -source_url = "https://msrg.utoronto.ca/publications/?page=2" +external_url = "https://doi.org/10.1109/icbc48266.2020.9169430" +abstract = "Enterprises and Governments, alike, are leveraging distributed ledger technologies to solve traditional problems across domains. They consider private blockchains such as Hyperledger Fabric as a safe bet for the obvious security and privacy reasons. However, the tools for software reliability are not yet matured. In this work, we propose FabricUnit, a unit testing framework for Hyperledger Fabric clients. FabricUnit identifies the safe methods that do not alter the state and re-uses the setup execution (deleting any stale data and reinitializes the data). Our experiment shows a reduction of approximately 30% in the tests execution time." +++ diff --git a/content/publications/fame-failure-aware-mixture-of-experts-for-message-level-log-anomaly-detection.md b/content/publications/fame-failure-aware-mixture-of-experts-for-message-level-log-anomaly-detection.md new file mode 100644 index 0000000..0058553 --- /dev/null +++ b/content/publications/fame-failure-aware-mixture-of-experts-for-message-level-log-anomaly-detection.md @@ -0,0 +1,10 @@ ++++ +title = "FAME: Failure-Aware Mixture-of-Experts for Message-Level Log Anomaly Detection" +year = 2026 +authors = ["Huanchi Wang", "Zihang Huang", "Yifang Tian", "Kristina Dzeparoska", "Hans-Arno Jacobsen", "Alberto Leon-Garcia"] +venue = "arXiv" +publication_type = "ArXiv Preprint" +research = ["distributed-machine-learning"] +external_url = "https://arxiv.org/abs/2605.22779" +abstract = "Production systems generate millions of log lines daily, yet most anomaly detectors operate at the session or window-level, flagging groups of lines rather than identifying the specific message responsible. This coarse granularity forces operators to inspect many routine lines per alert. Message-level detection offers finer granularity, but remains challenging. A single event template may correspond to both normal and anomalous messages, failures arise from heterogeneous subsystems, and line-level labeling at scale is impractical. Although large language models (LLMs) can reason over log semantics, applying them to every line is too costly for continuous monitoring. We present FAME (Failure-Aware Mixture-of-Experts), a label-efficient message-level mixture-of-experts framework that uses an LLM only once offline. We annotate at most K labeled lines per template to derive binary normal/anomaly indicators and representative examples. The LLM proposes a partition of templates into failure domains, and a certification step validates the proposal before training. FAME trains a lightweight router and domain experts that run on-premise and output anomaly predictions and failure-domain labels. On BGL, FAME achieves F1 = 98.16 at K = 100 reducing annotation effort by 76x and detects 97.7% of anomalies from unseen EventIDs. On Thunderbird, FAME reaches F1 = 99.95 with perfect recall." ++++ diff --git a/content/publications/federated-computing-in-electric-vehicles-to-predict-coolant-temperature.md b/content/publications/federated-computing-in-electric-vehicles-to-predict-coolant-temperature.md new file mode 100644 index 0000000..decb831 --- /dev/null +++ b/content/publications/federated-computing-in-electric-vehicles-to-predict-coolant-temperature.md @@ -0,0 +1,10 @@ ++++ +title = "Federated Computing in Electric Vehicles to Predict Coolant Temperature" +year = 2023 +authors = ["René Schwermer", "Ekin-Alp Bicer", "Pascal A. Schirmer", "Ruben Mayer", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 24th International Middleware Conference: Industrial Track" +publication_type = "Conference Paper" +research = ["distributed-machine-learning"] +external_url = "https://doi.org/10.1145/3626562.3626829" +abstract = "Reducing greenhouse gas emissions in mobility is paramount to achieving a carbon-neutral society. However, battery-electrical vehicles (BEV) introduce unique engineering challenges to protect expensive electrical components from overheating. A centralized architecture for model-driven predictions of coolant temperatures poses privacy and legal issues. Additionally, the applications in a vehicle compete for the available resources and must use them as sparingly as possible. Therefore, we introduce a new federated computing (FC) use case to help transform the mobility sector. We evaluate the performance of two FC approaches (linear regression and machine learning) on hardware and privacy metrics by leveraging a real-world dataset from BEVs. Our findings show trade-offs between hardware utilization and model accuracy. The linear regression model yields the best performance and prediction metrics. FC with ML shows up to 761 % variances when comparing vehicle-specific models with models trained with the entire fleet and clustering the data into velocity profiles partly improves prediction performance." ++++ diff --git a/content/publications/federated-computing-survey-on-building-blocks-extensions-and-systems.md b/content/publications/federated-computing-survey-on-building-blocks-extensions-and-systems.md new file mode 100644 index 0000000..23866e9 --- /dev/null +++ b/content/publications/federated-computing-survey-on-building-blocks-extensions-and-systems.md @@ -0,0 +1,10 @@ ++++ +title = "Federated Computing -- Survey on Building Blocks, Extensions and Systems" +year = 2024 +authors = ["René Schwermer", "Ruben Mayer", "Hans-Arno Jacobsen"] +venue = "arXiv" +publication_type = "ArXiv Preprint" +research = ["distributed-machine-learning"] +external_url = "https://arxiv.org/abs/2404.02779" +abstract = "In response to the increasing volume and sensitivity of data, traditional centralized computing models face challenges, such as data security breaches and regulatory hurdles. Federated Computing (FC) addresses these concerns by enabling collaborative processing without compromising individual data privacy. This is achieved through a decentralized network of devices, each retaining control over its data, while participating in collective computations. The motivation behind FC extends beyond technical considerations to encompass societal implications. As the need for responsible AI and ethical data practices intensifies, FC aligns with the principles of user empowerment and data sovereignty. FC comprises of Federated Learning (FL) and Federated Analytics (FA). FC systems became more complex over time and they currently lack a clear definition and taxonomy describing its moving pieces. Current surveys capture domain-specific FL use cases, describe individual components in an FC pipeline individually or decoupled from each other, or provide a quantitative overview of the number of published papers. This work surveys more than 150 papers to distill the underlying structure of FC systems with their basic building blocks, extensions, architecture, environment, and motivation. We capture FL and FA systems individually and point out unique difference between those two." ++++ diff --git a/content/publications/federated-fine-tuning-of-llms-on-the-very-edge.md b/content/publications/federated-fine-tuning-of-llms-on-the-very-edge.md index 8d31f8e..e7b8d60 100644 --- a/content/publications/federated-fine-tuning-of-llms-on-the-very-edge.md +++ b/content/publications/federated-fine-tuning-of-llms-on-the-very-edge.md @@ -1,11 +1,11 @@ +++ -title = "Federated Fine-tuning of LLMs on the Very Edge: The Good, the Bad, the Ugly" +title = "Federated Fine-Tuning of LLMs on the Very Edge: The Good, the Bad, the Ugly" year = 2024 -authors = ["Herbert Woisetschlaeger", "Alexander Erben", "Shiqiang Wang", "Bill Marino", "Nicholas D. Lane", "Ruben Mayer", "Hans-Arno Jacobsen"] +authors = ["Herbert Woisetschläger", "Alexander Erben", "Shiqiang Wang", "Ruben Mayer", "Hans-Arno Jacobsen"] venue = "Proceedings of the Eighth Workshop on Data Management for End-to-End Machine Learning" publication_type = "Conference Paper" research = ["distributed-machine-learning"] tags = ["federated-learning", "edge-systems", "llms"] -summary = "Conference paper on what it takes to fine-tune large language models through federated learning on extremely resource-constrained edge systems." -external_url = "https://dl.acm.org/doi/abs/10.1145/3650203.3663331" +external_url = "https://doi.org/10.1145/3650203.3663331" +abstract = "With the emergence of AI regulations, such as the EU AI Act, requirements for simple data lineage, enforcement of low data bias, and energy efficiency have become a priority for everyone offering AI services. Being pre-trained on versatile and a vast amount of data, large language models and foundation models (FMs) offer a good basis for building high-quality deep learning pipelines. Fine-tuning can further improve model performance on a specific downstream task, which requires orders of magnitude less data than pre-training. Often, access to high-quality and low-bias data for model fine-tuning is limited due to technical or regulatory requirements. Federated learning (FL), as a distributed and privacy-preserving technique, offers a well-suited approach to significantly expanding data access for model fine-tuning. Yet, this data is often located on the network edge, where energy, computational, and communication resources are significantly more limited than in data centers." +++ diff --git a/content/publications/federated-learning-and-ai-regulation-in-the-european-union-who-is-responsible-an-interdisciplinary-analysis.md b/content/publications/federated-learning-and-ai-regulation-in-the-european-union-who-is-responsible-an-interdisciplinary-analysis.md index 92586b9..dd6cec2 100644 --- a/content/publications/federated-learning-and-ai-regulation-in-the-european-union-who-is-responsible-an-interdisciplinary-analysis.md +++ b/content/publications/federated-learning-and-ai-regulation-in-the-european-union-who-is-responsible-an-interdisciplinary-analysis.md @@ -1,11 +1,11 @@ +++ -title = "Federated Learning and AI Regulation in the European Union: Who is Responsible? An Interdisciplinary Analysis" +title = "Federated Learning and AI Regulation in the European Union: Who is Responsible? – An Interdisciplinary Analysis" year = 2024 authors = ["Herbert Woisetschläger", "Simon Mertel", "Christoph Krönke", "Ruben Mayer", "Hans-Arno Jacobsen"] -venue = "arXiv" -publication_type = "ArXiv Preprint" +venue = "GenLaw Workshop at ICML 2024" +publication_type = "Workshop Paper" research = ["distributed-machine-learning"] tags = ["federated-learning"] -external_url = "https://arxiv.org/abs/2407.08105" -source_url = "https://msrg.utoronto.ca/publications/" +external_url = "https://blog.genlaw.org/pdfs/genlaw_icml2024/16.pdf" +abstract = "The European Union Artificial Intelligence Act mandates clear stakeholder responsibilities in developing and deploying machine learning applications to avoid substantial fines, prioritizing private and secure data processing with data remaining at its origin. Federated Learning (FL) enables the training of generative AI Models across data siloes, sharing only model parameters while improving data security. Since FL is a cooperative learning paradigm, clients and servers naturally share legal responsibility in the FL pipeline. Our work contributes to clarifying the roles of both parties, explains strategies for shifting responsibilities to the server operator, and points out open technical challenges that we must solve to improve FL's practical applicability under the EU AI Act." +++ diff --git a/content/publications/federated-learning-priorities-under-the-european-union-ai-act.md b/content/publications/federated-learning-priorities-under-the-european-union-ai-act.md index a275d50..99ccd00 100644 --- a/content/publications/federated-learning-priorities-under-the-european-union-ai-act.md +++ b/content/publications/federated-learning-priorities-under-the-european-union-ai-act.md @@ -1,16 +1,11 @@ +++ title = "Federated Learning Priorities Under the European Union Artificial Intelligence Act" year = 2024 -authors = ["Herbert Woisetschlaeger", "Alexander Erben", "Bill Marino", "Shiqiang Wang", "Nicholas D. Lane", "Ruben Mayer", "Hans-Arno Jacobsen"] -venue = "Technical Report" -publication_type = "Technical Report" +authors = ["Herbert Woisetschläger", "Alexander Erben", "Bill Marino", "Shiqiang Wang", "Nicholas D. Lane", "Ruben Mayer", "Hans-Arno Jacobsen"] +venue = "arXiv" +publication_type = "ArXiv Preprint" research = ["distributed-machine-learning"] tags = ["federated-learning", "ai-regulation", "systems-policy"] -summary = "Interdisciplinary analysis of how the EU AI Act may reshape priorities in federated learning research." external_url = "https://arxiv.org/abs/2402.05968" +abstract = "The age of AI regulation is upon us, with the European Union Artificial Intelligence Act (AI Act) leading the way. Our key inquiry is how this will affect Federated Learning (FL), whose starting point of prioritizing data privacy while performing ML fundamentally differs from that of centralized learning. We believe the AI Act and future regulations could be the missing catalyst that pushes FL toward mainstream adoption. However, this can only occur if the FL community reprioritizes its research focus. In our position paper, we perform a first-of-its-kind interdisciplinary analysis (legal and ML) of the impact the AI Act may have on FL and make a series of observations supporting our primary position through quantitative and qualitative analysis. We explore data governance issues and the concern for privacy. We establish new challenges regarding performance and energy efficiency within lifecycle monitoring. Taken together, our analysis suggests there is a sizable opportunity for FL to become a crucial component of AI Act-compliant ML systems and for the new regulation to drive the adoption of FL techniques in general. Most noteworthy are the opportunities to defend against data bias and enhance private and secure computation" +++ - -This position paper argues that AI regulation may become a real adoption driver -for federated learning, but only if the research community revisits its own -priorities. The paper connects legal requirements with systems concerns such as -privacy, data governance, efficiency, and lifecycle monitoring. diff --git a/content/publications/federated-office-plug-load-identification-for-building-management-systems.md b/content/publications/federated-office-plug-load-identification-for-building-management-systems.md new file mode 100644 index 0000000..dcc88ac --- /dev/null +++ b/content/publications/federated-office-plug-load-identification-for-building-management-systems.md @@ -0,0 +1,10 @@ ++++ +title = "Federated office plug-load identification for building management systems" +year = 2022 +authors = ["René Schwermer", "Jonas Buchberger", "Ruben Mayer", "Hans-Arno Jacobsen"] +venue = "Proceedings of the Thirteenth ACM International Conference on Future Energy Systems" +publication_type = "Conference Paper" +research = ["distributed-machine-learning"] +external_url = "https://doi.org/10.1145/3538637.3538845" +abstract = "Energy consumption in buildings is responsible for 40 % of the final energy consumption in the European Union and the United States of America. In addition to thermal energy, buildings require electricity for all kinds of appliances. Regulatory constraints such as energy labels aim at increasing the energy efficiency of large appliances such as fridges and washing machines. However, they only partially cover plug-loads. The amount of electricity consumption of unregulated plug-loads such as mobile phones, USB chargers and kettles is continuously increasing. For European households, their share of electricity consumption reached 25 % in 2018. Additional data about the plug-loads usage can help decrease the energy consumption of buildings by improving energy management systems, applying peak-shaving or demand-side management. People live and work in buildings, making such data privacy sensitive. Federated Learning (FL) helps to leverage these data without violating regulatory frameworks such as the General Data Protection Regulation. We use a high-frequency energy data set of office appliances (BLOND) to train four appliance classifiers (CNN, LSTM, ResNet and DenseNet). We investigate the effect of different data distributions (entire dataset, IID and non-IID) and training methods on four performance metrics (accuracy, F1 score, precision and recall). The results show that a non-IID setup decreases all performance metrics for some model architectures by 44 %. However, our LSTM model even with a non-IID labels achieves similar F1 scores compared to central training. Additionally, we show the importance of client selection in FL architectures to reduce the overall training time and we quantify the decrease in network traffic compared to a central training approach, the energy consumption and scalability." ++++ diff --git a/content/publications/fledge.md b/content/publications/fledge.md index 2e38f6b..b42f69a 100644 --- a/content/publications/fledge.md +++ b/content/publications/fledge.md @@ -1,17 +1,11 @@ +++ -title = "Fledge: Benchmarking Federated Machine Learning Applications in Edge Computing Systems" -year = 2023 -authors = ["Herbert Woisetschlaeger", "Alexander Isenko", "Ruben Mayer", "Hans-Arno Jacobsen"] -venue = "arXiv Preprint" -publication_type = "ArXiv Preprint" +title = "FLEdge: Benchmarking Federated Learning Applications in Edge Computing Systems" +year = 2024 +authors = ["Herbert Woisetschläger", "Alexander Erben", "Ruben Mayer", "Shiqiang Wang", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 25th International Middleware Conference" +publication_type = "Conference Paper" research = ["distributed-machine-learning"] tags = ["federated-learning", "benchmarking", "edge-systems"] -summary = "Benchmark for federated learning workloads in edge-computing environments with heterogeneous hardware and client failures." -external_url = "https://arxiv.org/abs/2306.05172" +external_url = "https://doi.org/10.1145/3652892.3700751" +abstract = "Federated Learning (FL) has become a viable technique for realizing privacy-enhancing distributed deep learning on the network edge. Heterogeneous hardware, unreliable client devices, and energy constraints often characterize edge computing systems. In this paper, we propose FLEdge, which complements existing FL benchmarks by enabling a systematic evaluation of client capabilities. We focus on computational and communication bottlenecks, client behavior, and data security implications. Our experiments with models varying from 14K to 80M trainable parameters are carried out on dedicated hardware with emulated network characteristics and client behavior. We find that state-of-the-art embedded hardware has significant memory bottlenecks, leading to 4× longer processing times than on modern data center GPUs." +++ - -Fledge addresses a common gap in federated-learning evaluation: most benchmarks -either simulate the system or assume data-center style deployments. This work -instead studies federated learning in more realistic edge settings, including -hardware diversity, energy efficiency, differential privacy, and heavy client -dropout. diff --git a/content/publications/flexible-caching-algorithms-for-video-content-distribution-networks.md b/content/publications/flexible-caching-algorithms-for-video-content-distribution-networks.md new file mode 100644 index 0000000..e3a7bdd --- /dev/null +++ b/content/publications/flexible-caching-algorithms-for-video-content-distribution-networks.md @@ -0,0 +1,10 @@ ++++ +title = "Flexible Caching Algorithms for Video Content Distribution Networks" +year = 2017 +authors = ["Kianoosh Mokhtarian", "Hans-Arno Jacobsen"] +venue = "IEEE/ACM Transactions on Networking" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.1109/tnet.2016.2621067" +abstract = "Global video content distribution networks (CDNs) serve a significant fraction of the entire Internet traffic. Effective caching at the edge is vital for the feasibility of these CDNs, which can otherwise incur substantial costs and overloads in the Internet. We analyze the challenges and requirements for content caching on the servers of these CDNs which cannot be addressed by standard solutions. We design multiple algorithms for this problem: a LRU-based baseline to address the requirements; a flexible ingress-efficient algorithm; an offline cache aware of future requests (greedy) to estimate the maximum efficiency we can expect from any online algorithm; an optimal offline cache (for limited scales); and an adaptive ingress control algorithm for reducing the server's peak upstream traffic. We use anonymized actual data from a global video CDN to evaluate the algorithms and draw conclusions on their suitability for different settings." ++++ diff --git a/content/publications/flexible-query-processor-on-fpgas.md b/content/publications/flexible-query-processor-on-fpgas.md new file mode 100644 index 0000000..5f939b8 --- /dev/null +++ b/content/publications/flexible-query-processor-on-fpgas.md @@ -0,0 +1,10 @@ ++++ +title = "Flexible Query Processor on FPGAs" +year = 2013 +authors = ["Mohammedreza Najafi", "Mohammad Sadoghi", "Hans-Arno Jacobsen"] +venue = "Proceedings of the VLDB Endowment" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.14778/2536274.2536303" +abstract = "In this work, we demonstrate Flexible Query Processor (FQP), an online reconfigurable event stream query processor. FQP is an FPGA-based query processor that supports select, project and join queries over event streams at line rate. While processing incoming events, FQP can accept new query expressions, a key distinguishing characteristic from related approaches employing FPGAs for acceleration. Our solution aims to address performance limitations experienced with general purpose processors needing to operate at line rate and lack of on the fly reconfigurability with custom designed hardware solutions on FPGAs." ++++ diff --git a/content/publications/forecasting-household-electricity-demand-with-complex-event-processing-insights-from-a-prototypical-solution.md b/content/publications/forecasting-household-electricity-demand-with-complex-event-processing-insights-from-a-prototypical-solution.md new file mode 100644 index 0000000..dd661f2 --- /dev/null +++ b/content/publications/forecasting-household-electricity-demand-with-complex-event-processing-insights-from-a-prototypical-solution.md @@ -0,0 +1,10 @@ ++++ +title = "Forecasting household electricity demand with complex event processing: insights from a prototypical solution" +year = 2013 +authors = ["Holger Ziekow", "Christoph Doblander", "Christoph Goebel", "Hans-Arno Jacobsen"] +venue = "Proceedings of the Industrial Track of the 13th ACM/IFIP/USENIX International Middleware Conference" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/2541596.2541598" +abstract = "The increasing use of renewable energy is leading to a paradigm shift in operating electrical grids. Production is moving away from centralized power plants to decentralized sources like solar panels and windmills. One consequence of this development is the need for managing supply and demand in local distribution grids in a \"smart\" way, which also implies the capability to forecast the demand for electric power closer to the end consumer and on shorter time scales than today. In this paper, we describe a system prototype for electricity demand forecasting based on highly disaggregated data from sensors deployed in homes and evaluate its performance both with respect to forecasting accuracy and ICT resource requirements. The data we use for our evaluation was collected in a pilot trial. Our system prototype combines complex event processing with state-of-the-art forecasting capabilities. For short-term forecasts, we observed average error reductions of up to 98 percentage points compared to average demand profiles. Our experiments also show the applicability of our approach at large scale. We were able to run the forecasting service for 1,000 households in parallel on one off-the-shelf server." ++++ diff --git a/content/publications/foundations-for-highly-available-content-based-publish-subscribe-overlays.md b/content/publications/foundations-for-highly-available-content-based-publish-subscribe-overlays.md new file mode 100644 index 0000000..b41c9ff --- /dev/null +++ b/content/publications/foundations-for-highly-available-content-based-publish-subscribe-overlays.md @@ -0,0 +1,10 @@ ++++ +title = "Foundations for Highly Available Content-Based Publish/Subscribe Overlays" +year = 2011 +authors = ["Young Yoon", "Vinod Muthusamy", "Hans-Arno Jacobsen"] +venue = "2011 31st International Conference on Distributed Computing Systems" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icdcs.2011.93" +abstract = "Content-based publish/subscribe overlays offer a scalable messaging substrate for various event-based distributed systems. In an enterprise environment where service level agreements(SLAs) are strictly enforced, maintaining high availability and efficiency of the broker overlay is critical. To support these requirements, a set of three primitive operations are proposed to allow arbitrary transformations of an overlay to an optima lone, and two additional primitives are developed to enable ondemand adjustments when there are permanent or transient failures. Both sets of primitive operations minimize disruption by preserving message delivery guarantees even as the overlay topology changes, requiring no overhead when the overlay is not being modified, operating on a fixed neighborhood of brokers regardless of the size of the overlay, and completing quickly under a variety of conditions." ++++ diff --git a/content/publications/fpga-topss-line-speed-event-processing-on-fpgas.md b/content/publications/fpga-topss-line-speed-event-processing-on-fpgas.md new file mode 100644 index 0000000..6b63611 --- /dev/null +++ b/content/publications/fpga-topss-line-speed-event-processing-on-fpgas.md @@ -0,0 +1,10 @@ ++++ +title = "fpga-ToPSS: line-speed event processing on fpgas" +year = 2011 +authors = ["Mohammad Sadoghi", "Harsh Singh", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 5th ACM international conference on Distributed event-based system" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/2002259.2002316" +abstract = "In this demo, we present fpga-ToPSS (a member of Toronto Publish/Subscribe System Family), an efficient event processing platform for high-frequency and low-latency algorithmic trading. Our event processing platform is built over reconfigurable hardware---FPGAs---to achieve line-rate processing. Furthermore, our event processing engine supports Boolean expression matching with an expressive predicate language that models complex financial strategies to autonomously mimic the buying and the selling of stocks based on real-time financial data." ++++ diff --git a/content/publications/fractalsort-high-precision-compressed-radix-sort-on-fpga.md b/content/publications/fractalsort-high-precision-compressed-radix-sort-on-fpga.md new file mode 100644 index 0000000..aa41d5e --- /dev/null +++ b/content/publications/fractalsort-high-precision-compressed-radix-sort-on-fpga.md @@ -0,0 +1,10 @@ ++++ +title = "FractalSort: High Precision Compressed Radix Sort on FPGA" +year = 2026 +authors = ["Michael Dang'ana", "Hans-Arno Jacobsen"] +venue = "IEEE Transactions on Computers" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.1109/tc.2026.3653702" +abstract = "State-of-the-art large data set high-precision sorting algorithms typically use hardware-accelerated radix sort. Advances in Dynamic Random Access Memory, Flash and High Bandwidth Memory (HBM) have enabled faster bandwidth intensive merge operations, where distribution-dependent data pre-processing techniques such as stochastic sampling bucketing offer alternatives impacted by increased data passes and vulnerability to data skew.This work addresses these limitations by introducing a compressed radix-sorting scheme for high-precision keys. Whereas radix sort histograms grow exponentially with precision, FractalSort guarantees bounded histogram size through the novel compression scheme which translates into smaller sorting circuits and reduced memory usage. Another key contribution is the novel optimized merge algorithm, which eliminates the need for data pre-processing and bucketing leading to higher bandwidth efficiency and reduced algorithm complexity. Using a tree-based recursive sorting architecture for space efficiency and low latency, the algorithm achieves fast sorting that exceeds the state-of-the-art on the CPU, FPGA, and GPU by 6x, 2.5x, and 3x bandwidth-adjusted throughput on 4GB to 2TB data sets. FractalSort is implemented on Xilinx Virtex UltraScale+ FPGA empirically demonstrating on-chip sorting at 20 Tb/s capable of memory-to-memory sorting of 32-bit keys at 3.2Tb/s using HBM." ++++ diff --git a/content/publications/fugue-online-elasticity-for-distributed-stateful-stream-processing.md b/content/publications/fugue-online-elasticity-for-distributed-stateful-stream-processing.md new file mode 100644 index 0000000..b902c53 --- /dev/null +++ b/content/publications/fugue-online-elasticity-for-distributed-stateful-stream-processing.md @@ -0,0 +1,10 @@ ++++ +title = "Fugue: Online Elasticity for Distributed Stateful Stream Processing" +year = 2026 +authors = ["Yuqiu Zhang", "Yunhao Mao", "Hans-Arno Jacobsen"] +venue = "Proceedings of the VLDB Endowment" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.14778/3836663.3836689" +abstract = "Stateful stream processing engines are critical for real-time analytics but lack efficient mechanisms for runtime elasticity. The dominant \"stop-the-world\" model, used by systems like Apache Flink, requires halting applications globally for a long time, while recent on-the-fly protocols introduce severe trade-offs: proactive approaches impose a continuous resource tax by constantly replicating state, and existing reactive solutions suffer from architectural complexity and external dependencies. This paper introduces Fugue, a novel, self-contained reactive protocol that provides seamless and resource-efficient elasticity. The core of Fugue is a two-phase design that combines a pre-emptive background state transfer with an atomic, lightweight barrier-based cutover. By moving the bulk of an operator's state off the critical path and unifying the final ownership transfer with the system's native exactly-once synchronization mechanism, Fugue guarantees correctness with minimal disruption and steady-state overhead. We implemented Fugue in Apache Flink and our evaluation on realistic benchmarks shows it reduces tail reconfiguration latency by up to 98.6% relative to native Flink while maintaining over 90% of peak throughput. Compared to reactive pull-based baselines, Fugue reduces end-to-end migration latency by up to 93.7%. Compared to proactive replication, it reaches comparable handover performance while avoiding continuous replication overhead. Together, these results demonstrate a strong combination of robustness, performance, and operational simplicity." ++++ diff --git a/content/publications/g-topss-fast-filtering-of-graph-based-metadata.md b/content/publications/g-topss-fast-filtering-of-graph-based-metadata.md new file mode 100644 index 0000000..8cf6e35 --- /dev/null +++ b/content/publications/g-topss-fast-filtering-of-graph-based-metadata.md @@ -0,0 +1,10 @@ ++++ +title = "G-ToPSS: fast filtering of graph-based metadata" +year = 2005 +authors = ["Milenko Petrovic", "Haifeng Liu", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 14th international conference on World Wide Web - WWW '05" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/1060745.1060824" +abstract = "RDF is increasingly being used to represent metadata. RDF Site Summary (RSS) is an application of RDF on the Web that has considerably grown in popularity. However, the way RSS systems operate today does not scale well. In this paper we introduce G-ToPSS, a scalable publish/subscribe system for selective information dissemination. G-ToPSS is particularly well suited for applications that deal with large-volume content distribution from diverse sources. RSS is an instance of the content distribution problem. G-ToPSS allows use of ontology as a way to provide additional information about the data. Furthermore, in this paper we show how G-ToPSS can support RDFS class taxonomies. We have implemented and experimentally evaluated G-ToPSS and we provide results in the paper demonstrating its scalability compared to alternatives." ++++ diff --git a/content/publications/gala-can-graph-augmented-large-language-model-agentic-workflows-elevate-root-cause-analysis.md b/content/publications/gala-can-graph-augmented-large-language-model-agentic-workflows-elevate-root-cause-analysis.md new file mode 100644 index 0000000..a643ed0 --- /dev/null +++ b/content/publications/gala-can-graph-augmented-large-language-model-agentic-workflows-elevate-root-cause-analysis.md @@ -0,0 +1,10 @@ ++++ +title = "GALA: Can Graph-Augmented Large Language Model Agentic Workflows Elevate Root Cause Analysis?" +year = 2025 +authors = ["Yifang Tian", "Yaming Liu", "Zichun Chong", "Zihang Huang", "Hans-Arno Jacobsen"] +venue = "arXiv" +publication_type = "ArXiv Preprint" +research = ["distributed-machine-learning", "data-management"] +external_url = "https://arxiv.org/abs/2508.12472" +abstract = "Root cause analysis (RCA) in microservice systems is challenging, requiring on-call engineers to rapidly diagnose failures across heterogeneous telemetry such as metrics, logs, and traces. Traditional RCA methods often focus on single modalities or merely rank suspect services, falling short of providing actionable diagnostic insights with remediation guidance. This paper introduces GALA, a novel multi-modal framework that combines statistical causal inference with LLM-driven iterative reasoning for enhanced RCA. Evaluated on an open-source benchmark, GALA achieves substantial improvements over state-of-the-art methods of up to 42.22% accuracy. Our novel human-guided LLM evaluation score shows GALA generates significantly more causally sound and actionable diagnostic outputs than existing methods. Through comprehensive experiments and a case study, we show that GALA bridges the gap between automated failure diagnosis and practical incident resolution by providing both accurate root cause identification and human-interpretable remediation guidance." ++++ diff --git a/content/publications/gala-graph-augmented-llm-agents-for-root-cause-analysis-and-incident-response-in-microservices.md b/content/publications/gala-graph-augmented-llm-agents-for-root-cause-analysis-and-incident-response-in-microservices.md new file mode 100644 index 0000000..93bedfc --- /dev/null +++ b/content/publications/gala-graph-augmented-llm-agents-for-root-cause-analysis-and-incident-response-in-microservices.md @@ -0,0 +1,10 @@ ++++ +title = "GALA: Graph-Augmented LLM Agents for Root Cause Analysis and Incident Response in Microservices" +year = 2026 +authors = ["Yifang Tian", "Yaming Liu", "Zichun Chong", "Zihang Huang", "Yiran Li", "Hans-Arno Jacobsen"] +venue = "arXiv" +publication_type = "ArXiv Preprint" +research = ["distributed-machine-learning", "data-management"] +external_url = "https://arxiv.org/abs/2608.08968" +abstract = "Microservice root cause analysis (RCA) requires correlating failures across heterogeneous telemetry within complex service dependency graphs. Existing methods often rely on a single telemetry modality; recent LLM-based approaches can suffer from unconstrained exploration and hallucination; and most systems stop at fault ranking without producing actionable incident response. We present GALA+, a graph-augmented LLM agentic framework centered on graph-guided investigation, which uses service dependencies to bound exploration and refine diagnosis through localized multi-modal evidence. For initial hypothesis generation, GALA+ combines complementary telemetry signals with STRIX, a novel trace- and graph-structure-aware scoring module. GALA+ then produces ranked diagnoses, incident summaries, and stratified action recommendations. We further introduce SURE-Score, a human-guided evaluation framework co-developed with industry SRE experts for assessing RCA-specific output quality beyond conventional text similarity metrics. On two microservice benchmarks, GALA+ consistently achieves the strongest overall results, surpassing the best LLM-based baseline by more than 25 percentage points in AC@1, while also receiving the highest ratings from both SURE-Score and independent human SRE evaluation." ++++ diff --git a/content/publications/generic-middleware-substrate-through-modelware.md b/content/publications/generic-middleware-substrate-through-modelware.md new file mode 100644 index 0000000..d6f96ca --- /dev/null +++ b/content/publications/generic-middleware-substrate-through-modelware.md @@ -0,0 +1,9 @@ ++++ +title = "Generic Middleware Substrate Through Modelware" +year = 2005 +authors = ["Charles Zhang", "Dapeng Gao", "Hans-Arno Jacobsen"] +venue = "Lecture Notes in Computer Science; Middleware 2005" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1007/11587552_16" ++++ diff --git a/content/publications/geo-distribution-of-flexible-business-processes.md b/content/publications/geo-distribution-of-flexible-business-processes.md index dd0c6a6..fd6d6ea 100644 --- a/content/publications/geo-distribution-of-flexible-business-processes.md +++ b/content/publications/geo-distribution-of-flexible-business-processes.md @@ -2,19 +2,12 @@ title = "Geo-Distribution of Flexible Business Processes over Publish/Subscribe Paradigm" slug = "geo-distribution-of-flexible-business-processes" year = 2016 -authors = ["Matthias Jergler", "Mohammad Sadoghi", "Hans-Arno Jacobsen"] -venue = "Proceedings of the 17th International Middleware Conference" +authors = ["Martin Jergler", "Mohammad Sadoghi", "Hans-Arno Jacobsen"] +venue = "Middleware" publication_type = "Conference Paper" research = ["data-management"] tags = ["business-process-management", "publish-subscribe", "geo-distribution"] -summary = "Presents a fully geo-distributed workflow engine for flexible business processes, built on the PADRES publish/subscribe system, with the context-aware mapping (CAM) that improves throughput and latency over a baseline mapping." -external_url = "https://dl.acm.org/doi/abs/10.1145/2988336.2988351" +external_url = "https://dl.acm.org/citation.cfm?id=2988351" related_datasets = ["geo-distribution-of-flexible-business-processes"] +abstract = "An increasing amount of business processes are inherently knowledge-intense and require ad-hoc decision making. Flexible modeling approaches such as the Case Management Model and Notation (CMMN) were designed to support such scenarios. At the same time, many processes involve participants and data from different organizations across the globe. Often, legal regulations such as data privacy render centralized execution engines impractical because data must be processed where it is collected. Instead, distributed approaches to coordinate process and data are necessary for supporting geo-scale execution. In this paper, we present a fully geo-distributed workflow engine that implements the core execution semantics of CMMN, the Guard-Stage-Milestone (GSM) meta-model, and supports locality of process data by distributing data and control-flow management over a loosely-coupled publish/subscribe infrastructure. We present a novel context-aware mapping (CAM) of GSM into Workflow Units (WFUs), representing the unit of distribution in our system. We have developed our distributed workflow execution engine over PADRES, an enterprise-grade event management system. Evaluation results show that our approach scales well with process size and degree of distribution and that CAM improves throughput and latency by up to 5X compared to the baseline mapping (BLM)." +++ - -Flexible business processes increasingly span participants and data across -organizations and geographies. This paper presents a fully geo-distributed workflow -engine that implements the Guard-Stage-Milestone meta-model on top of PADRES, with -a context-aware mapping (CAM) that improves throughput and latency by up to 5x -compared to a baseline mapping. The online appendix and companion material are -archived as a separate data set entry. diff --git a/content/publications/geospatial-event-analytics-leveraging-reactive-programming.md b/content/publications/geospatial-event-analytics-leveraging-reactive-programming.md new file mode 100644 index 0000000..4de33bc --- /dev/null +++ b/content/publications/geospatial-event-analytics-leveraging-reactive-programming.md @@ -0,0 +1,10 @@ ++++ +title = "Geospatial event analytics leveraging reactive programming" +year = 2015 +authors = ["Christoph Doblander", "Thomas Parsch", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 9th ACM International Conference on Distributed Event-Based Systems" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/2675743.2776757" +abstract = "In this paper, we present a solution to this year's DEBS Grand Challenge based on concepts from reactive systems. Reactive systems is a system architecture style with the following properties: Responsive, resilient, elastic, and message driven. When systems are built based on these properties, they tend to be more flexible, loosely-coupled, and scaleable. In this paper, we describe how to combine the operators given in the ReactiveX API to realize the individual challenge queries using asynchronous data-flows and evaluate the performance." ++++ diff --git a/content/publications/gpos-geospatially-aware-proof-of-stake.md b/content/publications/gpos-geospatially-aware-proof-of-stake.md new file mode 100644 index 0000000..e194df1 --- /dev/null +++ b/content/publications/gpos-geospatially-aware-proof-of-stake.md @@ -0,0 +1,10 @@ ++++ +title = "GPoS: Geospatially-aware Proof of Stake" +year = 2026 +authors = ["Shashank Motepalli", "Naman Garg", "Gengrui Zhang", "Hans-Arno Jacobsen"] +venue = "ACM Transactions on the Web" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.1145/3774629" +abstract = "Geospatial decentralization is essential for blockchains, ensuring regulatory resilience, robustness, and fairness. We empirically analyze five major Proof of Stake (PoS) blockchains—Aptos, Avalanche, Ethereum, Solana, and Sui—revealing that a few geographic regions dominate consensus voting power, resulting in limited geospatial decentralization. To address this, we propose Geospatially-aware Proof of Stake (GPoS), which integrates geospatial diversity with stake-based voting power. Experimental evaluation demonstrates an average 45% improvement in geospatial decentralization, as measured by the Gini coefficient of Eigenvector centrality, while incurring minimal performance overhead in BFT protocols, including HotStuff and CometBFT. These results demonstrate that GPoS can improve geospatial decentralization while, in our experiments, incurring minimal overhead to consensus performance." ++++ diff --git a/content/publications/gpx-matcher-a-generic-boolean-predicate-based-xpath-expression-matcher.md b/content/publications/gpx-matcher-a-generic-boolean-predicate-based-xpath-expression-matcher.md new file mode 100644 index 0000000..6645a7a --- /dev/null +++ b/content/publications/gpx-matcher-a-generic-boolean-predicate-based-xpath-expression-matcher.md @@ -0,0 +1,10 @@ ++++ +title = "GPX-matcher: a generic boolean predicate-based XPath expression matcher" +year = 2011 +authors = ["Mohammad Sadoghi", "Ioana Burcea", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 14th International Conference on Extending Database Technology" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/1951365.1951374" +abstract = "Content-based architectures for XML data dissemination are gaining increasing attention both in academia and industry. These dissemination networks are the building blocks of selective information dissemination applications which have wide applicability such as sharing and integrating information in both scientific and corporate domains. At the heart of these dissemination services is a fast engine for matching of an incoming XML message against stored XPath expressions to determine interested consumers for the message. To achieve the ultra-low response time, predominant in financial message processing, the XPath expression matching must be done efficiently. In this paper, we develop and evaluate a novel algorithm based on a unique encoding of XPath expressions and XML messages, unlike dominating automaton-based algorithms, for efficiently solving this matching problem. We demonstrate a matching time in the millisecond range for millions of XPath expressions which significantly outperforms state-of-the-art algorithms." ++++ diff --git a/content/publications/grand-challenge-real-time-soccer-analytics-leveraging-low-latency-complex-event-processing.md b/content/publications/grand-challenge-real-time-soccer-analytics-leveraging-low-latency-complex-event-processing.md new file mode 100644 index 0000000..25e337c --- /dev/null +++ b/content/publications/grand-challenge-real-time-soccer-analytics-leveraging-low-latency-complex-event-processing.md @@ -0,0 +1,10 @@ ++++ +title = "Grand challenge: real-time soccer analytics leveraging low-latency complex event processing" +year = 2013 +authors = ["Martin Jergler", "Christoph Doblander", "Mohammedreza Najafi", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 7th ACM international conference on Distributed event-based systems" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/2488222.2488280" +abstract = "In this paper, we present a real-time capable event-based system, which is tailored towards analytical query processing in the context of soccer games. The main challenge is to meet the application's strict real-time and low-latency requirements in face of streams of high-velocity sensor data. We describe a workflow-like architecture for query processing based on a publish/subscribe model. Queries are structured into computational tasks that are arranged sequentially and/or in parallel. Tasks are connected by preallocated ring buffers providing total event ordering and fast as well as decoupled event access. Our evaluation results show the effectiveness of the proposed system in terms of low-latency processing under real-time conditions. Speeding up the system by a factor of 50 compared to real-time introduces almost no latency overhead." ++++ diff --git a/content/publications/grand-challenge-the-bluebay-soccer-monitoring-engine.md b/content/publications/grand-challenge-the-bluebay-soccer-monitoring-engine.md new file mode 100644 index 0000000..d8c969e --- /dev/null +++ b/content/publications/grand-challenge-the-bluebay-soccer-monitoring-engine.md @@ -0,0 +1,10 @@ ++++ +title = "Grand challenge: the bluebay soccer monitoring engine" +year = 2013 +authors = ["Hans-Arno Jacobsen", "Kianoosh Mokhtarian", "Tilmann Rabl", "Mohammad Sadoghi", "Reza Sherafat Kazemzadeh", "Young Yoon", "Kaiwen Zhang"] +venue = "Proceedings of the 7th ACM international conference on Distributed event-based systems" +publication_type = "Conference Paper" +research = [] +external_url = "https://doi.org/10.1145/2488222.2488278" +abstract = "This paper presents the design and implementation of a custom-built event processing engine called BlueBay developed for live monitoring of soccer games. We experimentally evaluated our system using a real workload and report on its performance. Our results indicate that BlueBay achieves a throughput of up to 790k events per second, therefore processing the game's input sensor stream about 60 times faster than real-time. In addition to our custom implementation, we also investigated the applicability of off-the-shelf general-purpose event processing engines to address the soccer monitoring problem. This effort resulted in two additional and fully functional implementations based on Esper and Storm." ++++ diff --git a/content/publications/green-middleware.md b/content/publications/green-middleware.md new file mode 100644 index 0000000..c8d2112 --- /dev/null +++ b/content/publications/green-middleware.md @@ -0,0 +1,9 @@ ++++ +title = "Green Middleware" +year = 2011 +authors = ["Hans-Arno Jacobsen", "Vinod Muthusamy"] +venue = "Green IT: Technologies and Applications" +publication_type = "Book Chapter" +research = ["data-management"] +external_url = "https://doi.org/10.1007/978-3-642-22179-8_18" ++++ diff --git a/content/publications/green-resource-allocation-algorithms-for-publish-subscribe-systems.md b/content/publications/green-resource-allocation-algorithms-for-publish-subscribe-systems.md new file mode 100644 index 0000000..0bce9d6 --- /dev/null +++ b/content/publications/green-resource-allocation-algorithms-for-publish-subscribe-systems.md @@ -0,0 +1,10 @@ ++++ +title = "Green Resource Allocation Algorithms for Publish/Subscribe Systems" +year = 2011 +authors = ["Alex King Yeung Cheung", "Hans-Arno Jacobsen"] +venue = "2011 31st International Conference on Distributed Computing Systems" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icdcs.2011.82" +abstract = "A popular trend in large enterprises today is the adoption of green IT strategies that use resources as efficiently as possible to reduce IT operational costs. With the publish/subscribe middleware playing a vital role in seamlessly integrating applications at large enterprises including Google and Yahoo, our goal is to search for resource allocation algorithms that enable publish/subscribe systems to use system resources as efficiently as possible. To meet this goal, we develop methodologies that minimize system-wide message rates, broker load, hop count, and the number of allocated brokers, while maximizing the resource utilization of allocated brokers to achieve maximum efficiency. Our contributions consist of developing a bit vector supported resource allocation framework, designing and comparing four different classes with a total of ten variations of subscription allocation algorithms, and developing a recursive overlay construction algorithm. A compelling feature of our work is that it works under any arbitrary workload distribution and is independent of the publish/subscribe language, which makes it easily applicable to any topic and content-based publish/subscribe system. Experiments on a cluster testbed and a high performance computing platform show that our approach reduces the average broker message rate by up to 92% and the number of allocated brokers by up to 91%." ++++ diff --git a/content/publications/hardware-acceleration-landscape-for-distributed-real-time-analytics-virtues-and-limitations.md b/content/publications/hardware-acceleration-landscape-for-distributed-real-time-analytics-virtues-and-limitations.md new file mode 100644 index 0000000..ba6b25c --- /dev/null +++ b/content/publications/hardware-acceleration-landscape-for-distributed-real-time-analytics-virtues-and-limitations.md @@ -0,0 +1,10 @@ ++++ +title = "Hardware Acceleration Landscape for Distributed Real-Time Analytics: Virtues and Limitations" +year = 2017 +authors = ["Mohammadreza Najafi", "Kaiwen Zhang", "Mohammad Sadoghi", "Hans-Arno Jacobsen"] +venue = "2017 IEEE 37th International Conference on Distributed Computing Systems (ICDCS)" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icdcs.2017.194" +abstract = "We are witnessing a technological revolution with a broad impact ranging from daily life (e.g., personalized medicine and education) to industry (e.g., data-driven healthcare, commerce, agriculture, and mining). At the core of this transformation lies \"data\". This transformation is facilitated by embedded devices, collectively known as Internet of Things (IoT), which produce real-time feeds of sensor data which are collected and processed to produce a dynamic physical model used for optimized real-time decision making. At the infrastructure level, there is a need to develop a scalable architecture for processing massive volumes of present and historical data at an unprecedented velocity to support the IoT paradigm. To cope with such extreme scale, we argue for the need to revisit the hardware and software co-design landscape in light of two key technological advancements. First is the virtualization of computation and storage over highly distributed data centers spanning across continents. Second is the emergence of a variety of specialized hardware accelerators that complement traditional general-purpose processors. Further efforts are required to unify these two trends in order to harness the power of big data. In this paper, we present a formulation and characterization of the hardware acceleration landscape geared towards real-time analytics in the cloud. Our goal is to assist both researchers and practitioners navigating the newly revived field of software and hardware co-design for building next generation distributed systems. We further present a case study to explore software and hardware interplay for designing distributed real-time stream processing." ++++ diff --git a/content/publications/high-performance-stream-queries-in-scala.md b/content/publications/high-performance-stream-queries-in-scala.md new file mode 100644 index 0000000..ad64f03 --- /dev/null +++ b/content/publications/high-performance-stream-queries-in-scala.md @@ -0,0 +1,10 @@ ++++ +title = "High performance stream queries in scala" +year = 2015 +authors = ["Dantong Song", "Kaiwen Zhang", "Tilmann Rabl", "Prashanth Menon", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 9th ACM International Conference on Distributed Event-Based Systems" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/2675743.2776761" +abstract = "Traffic monitoring is an important stream processing application, which is highly dynamic and requires aggregation of spatially collocated data. Inspired by this, the DEBS 2015 Grand Challenge uses publicly available taxi transportation information to compute online the most frequent routes and most profitable areas. We describe our solution to the DEBS 2015 Grand Challenge, which can process events at a 10 ms latency and at a throughput of 114,000 events per second." ++++ diff --git a/content/publications/highly-available-content-based-publish-subscribe-via-gossiping.md b/content/publications/highly-available-content-based-publish-subscribe-via-gossiping.md new file mode 100644 index 0000000..be22493 --- /dev/null +++ b/content/publications/highly-available-content-based-publish-subscribe-via-gossiping.md @@ -0,0 +1,10 @@ ++++ +title = "Highly-available content-based publish/subscribe via gossiping" +year = 2016 +authors = ["Pooya Salehi", "Christoph Doblander", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 10th ACM International Conference on Distributed and Event-based Systems" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/2933267.2933303" +abstract = "Many publish/subscribe systems are based on a tree topology as their message dissemination overlay. However, in trees, even a single broker failure can cause delivery disruption. Hence, a repair mechanism is required, along with message retransmission to prevent message loss. During repair and recovery, the latency of message delivery can temporarily increase. To address this problem, we present an epidemic protocol to allow a content-based publish/subscribe system to keep delivering messages with low latency, while failed brokers are recovering. Using a broker similarity metric, which takes into account the content space and the overlay topology, we control and direct gossip messages around failed brokers. We compare our approach against a deterministic reliable publish/subscribe approach and an alternative epidemic approach. Based on our evaluations, we show that in our approach, the delivery ratio and latency of message deliveries are close to the deterministic approach, with up to 70% less message overhead than the alternative epidemic approach. Furthermore, our approach is able to provide a higher message delivery ratio than the deterministic alternative at high failure rates or when broker failures follow a non-uniform distribution." ++++ diff --git a/content/publications/historic-data-access-in-publish-subscribe.md b/content/publications/historic-data-access-in-publish-subscribe.md new file mode 100644 index 0000000..f010f79 --- /dev/null +++ b/content/publications/historic-data-access-in-publish-subscribe.md @@ -0,0 +1,10 @@ ++++ +title = "Historic data access in publish/subscribe" +year = 2007 +authors = ["Guoli Li", "Alex King Yeung Cheung", "Shuang Hou", "Songlin Hu", "Vinod Muthusamy", "R. Sherafat", "Alex Wun", "Hans-Arno Jacobsen", "Serge Mankovski"] +venue = "Proceedings of the 2007 inaugural international conference on Distributed event-based systems" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/1266894.1266908" +abstract = "We develop a content-based publish/subscribe platform, called PADRES, which is a distributed middleware platform with features inspired by the requirements of workflow management and business process execution. These features constitute original additions to publish/subscribe systems and include an expressive subscription language, historic, query-based data access, composite subscription processing, a rule-based matching and routing mechanism, and the support for the decentralized execution of service-oriented applications." ++++ diff --git a/content/publications/household-electricity-demand-forecasting-benchmarking-state-of-the-art-methods.md b/content/publications/household-electricity-demand-forecasting-benchmarking-state-of-the-art-methods.md new file mode 100644 index 0000000..b78547f --- /dev/null +++ b/content/publications/household-electricity-demand-forecasting-benchmarking-state-of-the-art-methods.md @@ -0,0 +1,10 @@ ++++ +title = "Household electricity demand forecasting: benchmarking state-of-the-art methods" +year = 2014 +authors = ["Andreas Veit", "Christoph Goebel", "Rohit Tidke", "Christoph Doblander", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 5th international conference on Future energy systems" +publication_type = "Conference Paper" +research = [] +external_url = "https://doi.org/10.1145/2602044.2602082" +abstract = "We benchmark state-of-the-art methods for forecasting electricity demand on the household level. Our evaluation is based on two data sets containing the power usage on the individual appliance level. Our results indicate that without further refinement the considered advanced state-of-the-art forecasting methods rarely beat corresponding persistence forecasts. Therefore, we also provide an exploration of promising directions for future research." ++++ diff --git a/content/publications/how-can-we-train-deep-learning-models-across-clouds-and-continents-an-experimental-study.md b/content/publications/how-can-we-train-deep-learning-models-across-clouds-and-continents-an-experimental-study.md new file mode 100644 index 0000000..611121c --- /dev/null +++ b/content/publications/how-can-we-train-deep-learning-models-across-clouds-and-continents-an-experimental-study.md @@ -0,0 +1,10 @@ ++++ +title = "How Can We Train Deep Learning Models Across Clouds and Continents? An Experimental Study" +year = 2024 +authors = ["Alexander Erben", "Ruben Mayer", "Hans-Arno Jacobsen"] +venue = "Proceedings of the VLDB Endowment" +publication_type = "Journal Article" +research = ["distributed-machine-learning"] +external_url = "https://doi.org/10.14778/3648160.3648165" +abstract = "This paper aims to answer the question: Can deep learning models be cost-efficiently trained on a global market of spot VMs spanning different data centers and cloud providers? To provide guidance, we extensively evaluate the cost and throughput implications of training in different zones, continents, and clouds for representative CV, NLP and ASR models. To expand the current training options further, we compare the scalability potential for hybrid-cloud scenarios by adding cloud resources to on-premise hardware to improve training throughput. Finally, we show how leveraging spot instance pricing enables a new cost-efficient way to train models with multiple cheap VMs, trumping both more centralized and powerful hardware and even on-demand cloud offerings at competitive prices." ++++ diff --git a/content/publications/how-does-stake-distribution-influence-consensus.md b/content/publications/how-does-stake-distribution-influence-consensus.md index c39334e..6b41c4f 100644 --- a/content/publications/how-does-stake-distribution-influence-consensus.md +++ b/content/publications/how-does-stake-distribution-influence-consensus.md @@ -6,6 +6,6 @@ venue = "2024 IEEE International Conference on Blockchain and Cryptocurrency (IC publication_type = "Conference Paper" research = ["data-management"] tags = ["blockchain", "consensus", "decentralization"] -summary = "Conference paper on how stake allocation shapes consensus behavior and decentralization outcomes in blockchain systems." -external_url = "https://ieeexplore.ieee.org/abstract/document/10634400/" +external_url = "https://doi.org/10.1109/icbc59979.2024.10634400" +abstract = "In the PoS blockchain landscape, the challenge of achieving full decentralization is often hindered by a disproportionate concentration of staked tokens among a few validators. This study analyses this challenge by first formalizing decentralization metrics for weighted consensus mechanisms. An empirical analysis across ten permissionless blockchains uncovers significant weight concentration among validators, underscoring the need for an equitable approach. To counter this, we introduce the Square Root Stake Weight (SRSW) model, which effectively recalibrates staking weight distribution. Our examination of the SRSW model demonstrates notable improvements in the decentralization metrics: the Gini index improves by $37.16 \\%$ on average, while Nakamoto coefficients for liveness and safety see mean enhancements of $101.04 \\%$ and $80.09 \\%$, respectively. This research is a pivotal step toward a more fair and equitable distribution of staking weight, advancing the decentralization in blockchain consensus mechanisms." +++ diff --git a/content/publications/how-reliable-are-streams-end-to-end-processing-guarantee-validation-and-performance-benchmarking-of-stream-processing-systems.md b/content/publications/how-reliable-are-streams-end-to-end-processing-guarantee-validation-and-performance-benchmarking-of-stream-processing-systems.md new file mode 100644 index 0000000..29fd9f6 --- /dev/null +++ b/content/publications/how-reliable-are-streams-end-to-end-processing-guarantee-validation-and-performance-benchmarking-of-stream-processing-systems.md @@ -0,0 +1,10 @@ ++++ +title = "How Reliable Are Streams? End-to-End Processing-Guarantee Validation and Performance Benchmarking of Stream Processing Systems" +year = 2024 +authors = ["Jawad Tahir", "Ruben Mayer", "Christoph Doblander", "Hans-Arno Jacobsen"] +venue = "Proceedings of the VLDB Endowment" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.14778/3712221.3712227" +abstract = "Stream processing systems (SPSs) provide processing guarantees to ensure reliability under failure. However, no related work exists that empirically validates these guarantees. In this paper, we present PGVal, a tool that can end-to-end validate guarantees of SPSs. Additionally, we introduce new metrics for SPSs, such as reliability, reliable throughput, and failure cost, in addition to a refined definition of latency that results in improved measurements. We benchmark three popular SPSs, namely Kafka Streams, Apache Storm , and Apache Flink. Our results show that the reliability of SPSs depends on many characteristics, such as data rate, data partitions, processing topology, and parallelism factor. An SPS configuration may not continue to provide reliable outputs when any of these characteristics vary. PGVal can also inject faults into SPSs to observe their impact on reliability and performance. We provide a comprehensive failure model for fault-tolerance benchmarking of SPSs and report on the impact of faults on the reliability and performance of SPSs. Our experiments show that SPSs' reliability and performance drop varies by fault. Lastly, we provide suggestions to increase the reliability and performance of these systems." ++++ diff --git a/content/publications/how-to-optimize-my-blockchain-a-multi-level-recommendation-approach.md b/content/publications/how-to-optimize-my-blockchain-a-multi-level-recommendation-approach.md new file mode 100644 index 0000000..0bdaeac --- /dev/null +++ b/content/publications/how-to-optimize-my-blockchain-a-multi-level-recommendation-approach.md @@ -0,0 +1,10 @@ ++++ +title = "How To Optimize My Blockchain? A Multi-Level Recommendation Approach" +year = 2023 +authors = ["Jeeta Ann Chacko", "Ruben Mayer", "Hans-Arno Jacobsen"] +venue = "Proceedings of the ACM on Management of Data" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.1145/3588704" +abstract = "Aside from the conception of new blockchain architectures, existing blockchain optimizations in the literature primarily focus on system or data-oriented optimizations within prevailing blockchains. However, since blockchains handle multiple aspects ranging from organizational governance to smart contract design, a holistic approach that encompasses all the different layers of a given blockchain system is required to ensure that all optimization opportunities are taken into consideration. In this vein, we define a multi-level optimization recommendation approach that identifies optimization opportunities within a blockchain at the system, data, and user level. Multiple metrics and attributes are derived from a blockchain log and nine optimization recommendations are formalized. We implement an automated optimization recommendation tool, BlockOptR, based on these concepts. The system is extensively evaluated with a wide range of workloads covering multiple real-world scenarios. After implementing the recommended optimizations, we observe an average of 20% improvement in the success rate of transactions and an average of 40% improvement in latency." ++++ diff --git a/content/publications/hybrid-context-inconsistency-resolution-for-context-aware-services.md b/content/publications/hybrid-context-inconsistency-resolution-for-context-aware-services.md new file mode 100644 index 0000000..bb7520c --- /dev/null +++ b/content/publications/hybrid-context-inconsistency-resolution-for-context-aware-services.md @@ -0,0 +1,10 @@ ++++ +title = "Hybrid context inconsistency resolution for context-aware services" +year = 2011 +authors = ["Chenhua Chen", "Chunyang Ye", "Hans-Arno Jacobsen"] +venue = "2011 IEEE International Conference on Pervasive Computing and Communications (PerCom)" +publication_type = "Conference Paper" +research = [] +external_url = "https://doi.org/10.1109/percom.2011.5767574" +abstract = "Context-aware applications automatically adapt their behavior according to environmental conditions, also known as contexts. However, in practice contexts are often inaccurate, noisy or even inconsistent (e.g., two RFID readers may report different numbers for the same set of goods processed). These kinds of problematic contexts may cause context-aware applications to behave abnormally or even fail. It is thus desirable to detect and resolve context inconsistency. In this paper, we propose a hybrid approach to detect problematic contexts and resolve resulting context inconsistencies with the help of context-aware application semantics. By combining low-level context inconsistency resolution with high-level application error recovery, our approach can resolve the inconsistent contexts more effectively. Moreover, error recovery cost for context-aware applications is reduced. Our experimental results show that our approach outperforms existing approaches in terms of more accurate inconsistency resolution and less error recovery cost." ++++ diff --git a/content/publications/hybrid-edge-partitioner-partitioning-large-power-law-graphs-under-memory-constraints.md b/content/publications/hybrid-edge-partitioner-partitioning-large-power-law-graphs-under-memory-constraints.md new file mode 100644 index 0000000..34fff5a --- /dev/null +++ b/content/publications/hybrid-edge-partitioner-partitioning-large-power-law-graphs-under-memory-constraints.md @@ -0,0 +1,10 @@ ++++ +title = "Hybrid Edge Partitioner: Partitioning Large Power-Law Graphs under Memory Constraints" +year = 2021 +authors = ["Ruben Mayer", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 2021 International Conference on Management of Data" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/3448016.3457300" +abstract = "Distributed systems that manage and process graph-structured data internally solve a graph partitioning problem to minimize their communication overhead and query run-time. Besides computational complexity---optimal graph partitioning is NP-hard---another important consideration is the memory overhead. Real-world graphs often have an immense size, such that loading the complete graph into memory for partitioning is not economical or feasible. Currently, the common approach to reduce memory overhead is to rely on streaming partitioning algorithms. While the latest streaming algorithms lead to reasonable partitioning quality on some graphs, they are still not completely competitive to in-memory partitioners. In this paper, we propose a new system, Hybrid Edge Partitioner (HEP), that can partition graphs that fit partly into memory while yielding a high partitioning quality. HEP can flexibly adapt its memory overhead by separating the edge set of the graph into two sub-sets. One sub-set is partitioned by NE++, a novel, efficient in-memory algorithm, while the other sub-set is partitioned by a streaming approach. Our evaluations on large real-world graphs show that in many cases, HEP outperforms both in-memory partitioning and streaming partitioning at the same time. Hence, HEP is an attractive alternative to existing solutions that cannot fine-tune their memory overheads. Finally, we show that using HEP, we achieve a significant speedup of distributed graph processing jobs on Spark/GraphX compared to state-of-the-art partitioning algorithms." ++++ diff --git a/content/publications/hyscale-hybrid-and-network-scaling-of-dockerized-microservices-in-cloud-data-centres.md b/content/publications/hyscale-hybrid-and-network-scaling-of-dockerized-microservices-in-cloud-data-centres.md new file mode 100644 index 0000000..20c83ea --- /dev/null +++ b/content/publications/hyscale-hybrid-and-network-scaling-of-dockerized-microservices-in-cloud-data-centres.md @@ -0,0 +1,10 @@ ++++ +title = "HyScale: Hybrid and Network Scaling of Dockerized Microservices in Cloud Data Centres" +year = 2019 +authors = ["Anthony Kwan", "Jonathon Wong", "Hans-Arno Jacobsen", "Vinod Muthusamy"] +venue = "2019 IEEE 39th International Conference on Distributed Computing Systems (ICDCS)" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icdcs.2019.00017" +abstract = "When designing modern software, care must be taken to allow for applications to scale based on the demands of its users while still accommodating flexibility in development. Recently, microservices architectures have garnered the attention of many organizations-providing higher levels of scalability, availability, and fault isolation. Many organizations choose to host their microservices architectures in cloud data centres to offset costs. Incidentally, data centres become over-encumbered during peak usage hours and underutilized during off-peak hours. Traditional microservice scaling methods perform either horizontal or vertical scaling exclusively. When used in combination, however, these methods offer complementary benefits and compensate for each other's deficiencies. To leverage the high availability of horizontal scaling and the fine-grained resource control of vertical scaling, we developed two novel hybrid autoscaling algorithms and a dedicated network scaling algorithm and benchmarked them against Google's popular Kubernetes horizontal autoscaling algorithm. Results indicated up to 1.49x speedups in response times for our hybrid algorithms, and 1.69x speedups for our network algorithm under high-burst network loads." ++++ diff --git a/content/publications/i-know-what-you-mean-semantic-issues-in-internet-scale-publish-subscribe-systems.md b/content/publications/i-know-what-you-mean-semantic-issues-in-internet-scale-publish-subscribe-systems.md new file mode 100644 index 0000000..c308cb9 --- /dev/null +++ b/content/publications/i-know-what-you-mean-semantic-issues-in-internet-scale-publish-subscribe-systems.md @@ -0,0 +1,10 @@ ++++ +title = "I know what you mean: semantic issues in Internet-scale publish/subscribe systems" +year = 2003 +authors = ["Ioana Burcea", "Milenko Petrovic", "Hans-Arno Jacobsen"] +venue = "SWDB" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://dblp.org/rec/conf/semweb/BurceaPJ03" +abstract = "In recent years, the amount of information on the Internet has increased exponentially developing great interest in selective information dissemination systems. The publish/subscribe paradigm is particularly suited for designing systems for routing information and requests according to their content throughout wide-area network of brokers. Current publish/subscribe systems use limited syntax-based content routing but since publishers and subscribers are anonymous and decoupled in time, space and location, often over wide-area network boundary, they do not necessarily speak the same language. Consequently, adding semantics to current publish/subscribe systems is important. In this paper we identify and examine the issues in developing semantic-based content routing for publish/subscribe broker networks." ++++ diff --git a/content/publications/i13dr-a-distributed-real-time-demand-response-infrastructure-for-integrating-renewable-energy-resources.md b/content/publications/i13dr-a-distributed-real-time-demand-response-infrastructure-for-integrating-renewable-energy-resources.md new file mode 100644 index 0000000..6ad7011 --- /dev/null +++ b/content/publications/i13dr-a-distributed-real-time-demand-response-infrastructure-for-integrating-renewable-energy-resources.md @@ -0,0 +1,10 @@ ++++ +title = "i13DR: A Distributed Real-Time Demand Response Infrastructure for Integrating Renewable Energy Resources" +year = 2018 +authors = ["Pezhman Nasirifard", "José Rivera", "Martin Jergler", "Hans-Arno Jacobsen"] +venue = "Proceedings of the Ninth International Conference on Future Energy Systems" +publication_type = "Conference Paper" +research = [] +external_url = "https://doi.org/10.1145/3208903.3212053" +abstract = "With the ongoing integration of Renewable Energy Sources (RES), the complexity of power grids is increasing. Due to the fluctuating nature of RES, ensuring the reliability of power grids can be challenging. One possible approach for addressing these challenges is Demand Response (DR) which is described as matching the demand for electrical energy according to the changes and the availability of supply. However, implementing a DR system to monitor and control a broad set of electrical appliances in real-time introduces several new complications including ensuring reliability and financial feasibility of the system. In this work, we address these issues by designing and implementing a distributed real-time DR infrastructure for laptops, which estimates and controls the power consumption of a network of connected laptops in response to the fast irregular changes of RES. The result of our field experiments confirms that our system successfully schedules and executes rapid and effective DR events. However, the accuracy of estimated power consumption of all participating laptops is relatively low, directly caused by our software-based approach." ++++ diff --git a/content/publications/i13dr-a-real-time-demand-response-infrastructure-for-integrating-renewable-energy-resources.md b/content/publications/i13dr-a-real-time-demand-response-infrastructure-for-integrating-renewable-energy-resources.md new file mode 100644 index 0000000..de59529 --- /dev/null +++ b/content/publications/i13dr-a-real-time-demand-response-infrastructure-for-integrating-renewable-energy-resources.md @@ -0,0 +1,10 @@ ++++ +title = "i13DR: A Real-Time Demand Response Infrastructure for Integrating Renewable Energy Resources" +year = 2022 +authors = ["Pezhman Nasirifard", "Hans-Arno Jacobsen"] +venue = "arXiv" +publication_type = "ArXiv Preprint" +research = [] +external_url = "https://arxiv.org/abs/2210.07789" +abstract = "With the ongoing integration of Renewable Energy Sources (RES), the complexity of power grids is increasing. Due to the fluctuating nature of RES, ensuring the reliability of power grids can be challenging. One possible approach for addressing these challenges is Demand Response (DR) which is described as matching the demand for electrical energy according to the changes and the availability of supply. However, implementing a DR system to monitor and control a broad set of electrical appliances in real-time introduces several new complications, including ensuring the reliability and financial feasibility of the system. In this work, we address these issues by designing and implementing a distributed real-time DR infrastructure for laptops, which estimates and controls the power consumption of a network of connected laptops in response to the fast, irregular changes of RES. Furthermore, since our approach is entirely software-based, we dramatically reduce the initial costs of the demand side participants. The result of our field experiments confirms that our system successfully schedules and executes rapid and effective DR events. However, the accuracy of the estimated power consumption of all participating laptops is relatively low, directly caused by our software-based approach." ++++ diff --git a/content/publications/incremental-topology-transformation-for-publish-subscribe-systems-using-integer-programming.md b/content/publications/incremental-topology-transformation-for-publish-subscribe-systems-using-integer-programming.md new file mode 100644 index 0000000..a834923 --- /dev/null +++ b/content/publications/incremental-topology-transformation-for-publish-subscribe-systems-using-integer-programming.md @@ -0,0 +1,10 @@ ++++ +title = "Incremental Topology Transformation for Publish/Subscribe Systems Using Integer Programming" +year = 2017 +authors = ["Pooya Salehi", "Kaiwen Zhang", "Hans-Arno Jacobsen"] +venue = "2017 IEEE 37th International Conference on Distributed Computing Systems (ICDCS)" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icdcs.2017.17" +abstract = "Distributed overlay-based publish/subscribe systems provide a selective, scalable, and decentralized approach to data dissemination. Due to the dynamic communication flows between data producers and consumers, the overlay topology of such systems can become inefficient over time and therefore requires adaptation to the existing load. Existing studies propose algorithms to design overlay topologies which are optimized for specific workloads. However, the problem of generating a plan to incrementally transform the current topology to an optimized one has been largely ignored. In this paper, we present IPITT, an approach based on integer programming for the incremental topology transformation (ITT) problem. Given the current topology and a target topology, IPITT generates a transformation plan with a minimal number of steps in order to lessen service disruption. Furthermore, we introduce a plan execution mechanism and evaluate our approach on an existing publish/subscribe system. Based on our evaluation, IPITT can reduce plan computation time by a factor of 10 and generates plans with an execution time up to 55% shorter than those of existing approaches." ++++ diff --git a/content/publications/inference-of-distribution-grids-based-on-crowdsourced-grid-data-and-drone-imagery.md b/content/publications/inference-of-distribution-grids-based-on-crowdsourced-grid-data-and-drone-imagery.md new file mode 100644 index 0000000..e0ad791 --- /dev/null +++ b/content/publications/inference-of-distribution-grids-based-on-crowdsourced-grid-data-and-drone-imagery.md @@ -0,0 +1,10 @@ ++++ +title = "Inference of Distribution Grids Based on Crowdsourced Grid Data and Drone Imagery" +year = 2022 +authors = ["Hans-Arno Jacobsen", "Pezhman Nasirifard", "José Rivera", "Prerona Ray Baruah"] +venue = "IEEE Transactions on Sustainable Computing" +publication_type = "Journal Article" +research = ["distributed-machine-learning"] +external_url = "https://doi.org/10.1109/tsusc.2019.2961763" +abstract = "Distribution System Operators (DSOs) face several challenges in managing comprehensive and up-to-date models of distribution grids. To address these problems, we propose a crowdsourcing framework for collecting grid devices. We also provide an inference approach for generating topological models of the distribution grids. Since distribution cables are often underground, we use spatial data analytics on the collected data in combination with other open data sources to infer the topology of the distribution grid. Additionally, to increase the quality of crowdsourced data, we propose a cost-effective approach for collecting and detecting grid elements in urban areas using commercial drones with an RGB camera. To evaluate our approach, we organized a crowdsourcing campaign to map and infer a district in Munich, Germany. The results are compared with the ground truth of the distribution system operator. Our results report a precision of up to 82 percent and a recall of up to 65 percent for the correctly crowdsourced grid devices. We also observe that the inferred models achieve a power length accuracy of 88 percent compared to the ground truth. We evaluated the detection of solar panels from aerial imagery by conducting field experiments, showing precision and recall levels of 68 and 69 percent, respectively." ++++ diff --git a/content/publications/infrastructure-free-content-based-publish-subscribe.md b/content/publications/infrastructure-free-content-based-publish-subscribe.md new file mode 100644 index 0000000..51600d1 --- /dev/null +++ b/content/publications/infrastructure-free-content-based-publish-subscribe.md @@ -0,0 +1,10 @@ ++++ +title = "Infrastructure-Free Content-Based Publish/Subscribe" +year = 2014 +authors = ["Vinod Muthusamy", "Hans-Arno Jacobsen"] +venue = "IEEE/ACM Transactions on Networking" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.1109/tnet.2013.2282159" +abstract = "Peer-to-peer (P2P) networks can offer benefits to distributed content-based publish/subscribe data dissemination systems. In particular, since a P2P network's aggregate resources grow as the number of participants increases, scalability can be achieved using no infrastructure other than the participants' own resources. This paper proposes algorithms for supporting content-based publish/subscribe in which subscriptions can specify a range of interest and publications a range of values. The algorithms are built over a distributed hash table abstraction and are completely decentralized. Load balance is addressed by subscription delegation away from overloaded peers and a bottom-up tree search technique that avoids root hotspots. Furthermore, fault tolerance is achieved with a lightweight replication scheme that quickly detects and recovers from faults. Experimental results support the scalability and fault-tolerance properties of the algorithms: For example, doubling the number of subscriptions does not double internal system messages, and even the simultaneous failure of 20% of the peers in the system requires less than 2 min to fully recover." ++++ diff --git a/content/publications/just-can-t-get-enough-synthesizing-big-data.md b/content/publications/just-can-t-get-enough-synthesizing-big-data.md new file mode 100644 index 0000000..caa16f3 --- /dev/null +++ b/content/publications/just-can-t-get-enough-synthesizing-big-data.md @@ -0,0 +1,10 @@ ++++ +title = "Just can't get enough: Synthesizing Big Data" +year = 2015 +authors = ["Tilmann Rabl", "Manuel Danisch", "Michael Frank", "Sebastian Schindler", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 2015 ACM SIGMOD International Conference on Management of Data" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/2723372.2735378" +abstract = "With the rapidly decreasing prices for storage and storage systems ever larger data sets become economical. While only few years ago only successful transactions would be recorded in sales systems, today every user interaction will be stored for ever deeper analysis and richer user modeling. This has led to the development of big data systems, which offer high scalability and novel forms of analysis. Due to the rapid development and ever increasing variety of the big data landscape, there is a pressing need for tools for testing and benchmarking." ++++ diff --git a/content/publications/krysha-cost-efficient-resource-orchestration-for-geo-distributed-serverless-microservices.md b/content/publications/krysha-cost-efficient-resource-orchestration-for-geo-distributed-serverless-microservices.md new file mode 100644 index 0000000..5a714a2 --- /dev/null +++ b/content/publications/krysha-cost-efficient-resource-orchestration-for-geo-distributed-serverless-microservices.md @@ -0,0 +1,10 @@ ++++ +title = "Krysha: Cost-Efficient Resource Orchestration for Geo-Distributed Serverless Microservices" +year = 2026 +authors = ["Yuqiu Zhang", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 35th International Symposium on High-Performance Parallel and Distributed Computing" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/3806645.3807589" +abstract = "The convergence of microservice architectures and serverless computing promises an elastic and cost-efficient model for modern cloud applications that often span multiple geo-distributed regions. However, prevailing serverless orchestrators that prioritize resource utilization or simple cold-start mitigation often prove suboptimal concerning SLO compliance and cost-efficiency in this emerging use case. In this paper, we present Krysha, an adaptive orchestration framework that jointly optimizes function scheduling and resource allocation for geo-distributed serverless microservices. Krysha employs a novel bi-level scheduling strategy: global-level early-binding to regions for fast function dispersion, coupled with regional-level late-binding to compute nodes for optimized resource use and cost. Moreover, Krysha achieves fine-grained resource allocation by decoupling CPU and memory provisioning and applying in-place vertical scaling on individual function instances. These capabilities are guided by a comprehensive cost model and practical online optimization techniques. Our extensive evaluation shows that Krysha can achieve up to 74.7% cost savings in scaled deployments compared to state-of-the-art alternatives while maintaining SLO requirements." ++++ diff --git a/content/publications/ksurf-attention-kalman-filter-and-principal-component-analysis-for-prediction-under-highly-variable-cloud-workloads.md b/content/publications/ksurf-attention-kalman-filter-and-principal-component-analysis-for-prediction-under-highly-variable-cloud-workloads.md new file mode 100644 index 0000000..36555b3 --- /dev/null +++ b/content/publications/ksurf-attention-kalman-filter-and-principal-component-analysis-for-prediction-under-highly-variable-cloud-workloads.md @@ -0,0 +1,10 @@ ++++ +title = "Ksurf: Attention Kalman Filter and Principal Component Analysis for Prediction under Highly Variable Cloud Workloads" +year = 2024 +authors = ["Michael Dang'ana", "Hans-Arno Jacobsen"] +venue = "2024 11th International Conference on Electrical Engineering, Computer Science and Informatics (EECSI)" +publication_type = "Conference Paper" +research = [] +external_url = "https://doi.org/10.1109/eecsi63442.2024.10776180" +abstract = "Resource estimation and workload forecasting are critical in cloud data centers. Complexity in the cloud provider environment due to varying numbers of virtual machines introduces high variability in workloads and resource usage, making estimations problematic using state-of-the-art models that fail to deal with nonlinear characteristics. High measurement noise and variance affect the estimation of resource metrics of cloud systems across packet networks influenced by external dynamics. An ideal solution to these problems is the Kalman filter, a variance-minimizing estimator, ideal for highly variable data with Gaussian state space noise such as internet workloads. This work provides a new solution by making these contributions: i) it introduces a novel Kalman filter estimator using principal component analysis and an attention mechanism, ii) it evaluates the scheme on a Google Cloud benchmark comparing it to the state-of-the-art Bi-directional Grid Long Short- Term Memory network model on prediction tasks and iii) demonstrates real-time performance through a control task using a cloud-based messaging system with predictive auto-scaling. The new scheme improves prediction accuracy by 37% over state-of-the-art Kalman filters in prediction tasks, reduces the time series prediction error of the neural network model by over 40%, and improves Apache Kafka workload-based scaling stability by 58%." ++++ diff --git a/content/publications/ksurf-attention-kalman-filter-for-prediction-under-highly-variable-cloud-workloads.md b/content/publications/ksurf-attention-kalman-filter-for-prediction-under-highly-variable-cloud-workloads.md new file mode 100644 index 0000000..7ad6d01 --- /dev/null +++ b/content/publications/ksurf-attention-kalman-filter-for-prediction-under-highly-variable-cloud-workloads.md @@ -0,0 +1,10 @@ ++++ +title = "Ksurf+: Attention Kalman Filter for Prediction Under Highly Variable Cloud Workloads" +year = 2025 +authors = ["Michael Dang'ana", "Hans-Arno Jacobsen"] +venue = "IEEE Transactions on Cloud Computing" +publication_type = "Journal Article" +research = [] +external_url = "https://doi.org/10.1109/tcc.2025.3605828" +abstract = "Resource estimation and workload forecasting are critical in cloud data centers. Complexity in the cloud provider environment due to varying numbers of virtual machines introduces high variability in workloads and resource usage, making estimations problematic using state-of-the-art models that fail to deal with nonlinear characteristics. High measurement noise and variance affect the estimation of resource metrics of cloud systems across packet networks influenced by unknown external dynamics. An ideal solution to these problems is the Kalman filter, a variance-minimizing estimator, ideal for highly variable data. This work introduces Ksurf+, a novel Kalman filter estimator using selective principal component analysis and an attention mechanism for enhanced short-horizon prediction. Ksurf+ improves prediction accuracy by 37% over state-of-the-art Kalman filters in prediction tasks, reduces the time series prediction error of the state-of-the-art Bi-directional Grid Long Short-Term Memory neural network by over 40%, improves Kafka workload-based scaling stability by 58%, reduces Kafka queue size and lowers Kubernetes worker pod CPU usage by 11.6% on the $VarBench$ benchmark." ++++ diff --git a/content/publications/ksurf-drone-attention-kalman-filter-for-contextual-bandit-optimization-in-cloud-resource-allocation.md b/content/publications/ksurf-drone-attention-kalman-filter-for-contextual-bandit-optimization-in-cloud-resource-allocation.md new file mode 100644 index 0000000..680f0c5 --- /dev/null +++ b/content/publications/ksurf-drone-attention-kalman-filter-for-contextual-bandit-optimization-in-cloud-resource-allocation.md @@ -0,0 +1,10 @@ ++++ +title = "Ksurf-Drone: Attention Kalman Filter for Contextual Bandit Optimization in Cloud Resource Allocation" +year = 2026 +authors = ["Michael Dang'ana", "Yuqiu Zhang", "Hans-Arno Jacobsen"] +venue = "IEEE Transactions on Cloud Computing" +publication_type = "Journal Article" +research = [] +external_url = "https://doi.org/10.1109/tcc.2026.3653558" +abstract = "Resource orchestration and configuration parameter search are key concerns for container-based infrastructure in cloud data centers. Large configuration search space and cloud uncertainties are often mitigated using contextual bandit techniques for resource orchestration including the state-of-the-art Drone orchestrator. Complexity in the cloud provider environment due to varying numbers of virtual machines introduces variability in workloads and resource metrics, making orchestration decisions less accurate due to increased nonlinearity and noise. Ksurf, a state-of-the-art variance-minimizing estimator method ideal for highly variable cloud data, enables optimal resource estimation under conditions of high cloud variability. This work evaluates the performance of Ksurf on estimation-based resource orchestration tasks involving highly variable workloads when employed as a contextual multi-armed bandit objective function model for cloud scenarios using Drone. Ksurf enables significantly lower latency variance of over $40\\%$ at p95 and p99, demonstrates significant reduction in CPU and master node memory usage on Kubernetes, resulting in a $7\\%$ cost savings in average worker pod count on $VarBench$ Kubernetes benchmark." ++++ diff --git a/content/publications/lifting-the-fog-of-uncertainties.md b/content/publications/lifting-the-fog-of-uncertainties.md index 55f3f52..5a65bd4 100644 --- a/content/publications/lifting-the-fog-of-uncertainties.md +++ b/content/publications/lifting-the-fog-of-uncertainties.md @@ -6,6 +6,6 @@ venue = "Proceedings of the 2023 ACM Symposium on Cloud Computing" publication_type = "Conference Paper" research = ["data-management"] tags = ["cloud-systems", "resource-orchestration", "containers"] -summary = "Conference paper on dynamic resource orchestration for containerized cloud environments." -external_url = "https://dl.acm.org/doi/abs/10.1145/3620678.3624646" +external_url = "https://doi.org/10.1145/3620678.3624646" +abstract = "The advances in virtualization technologies have sparked a growing transition from virtual machine (VM)-based to container-based infrastructure for cloud computing. From the resource orchestration perspective, containers' lightweight and highly configurable nature not only enables opportunities for more optimized strategies, but also poses greater challenges due to additional uncertainties and a larger configuration parameter search space. Towards this end, we propose Drone, a resource orchestration framework that adaptively configures resource parameters to improve application performance and reduce operational cost in the presence of cloud uncertainties. Built on Contextual Bandit techniques, Drone is able to achieve a balance between performance and resource cost on public clouds, and optimize performance on private clouds where a hard resource constraint is present. We show that our algorithms can achieve sub-linear growth in cumulative regret, a theoretically sound convergence guarantee, and our extensive experiments show that Drone achieves an up to 45% performance improvement and a 20% resource footprint reduction across batch processing jobs and microservice workloads." +++ diff --git a/content/publications/linking-goals-to-aspects.md b/content/publications/linking-goals-to-aspects.md new file mode 100644 index 0000000..1a65549 --- /dev/null +++ b/content/publications/linking-goals-to-aspects.md @@ -0,0 +1,10 @@ ++++ +title = "Linking goals to aspects" +year = 2005 +authors = ["Charles Zhang", "Hans-Arno Jacobsen", "Yijun Yu"] +venue = "Early Aspects: Aspect-Oriented Requirements Engineering and Architecture Design" +publication_type = "Workshop Paper" +research = [] +external_url = "https://oro.open.ac.uk/33844/" +abstract = "In RE models such as goal-oriented models, a complex sys-tem is directly described in terms of its purposes, which makes its functionality much easier to understand and to reason as compared to code-level implementations. Part of the difficulty in maintaining a stronger correspondence between requirements and code is possibly due to the suffi-cient modularization capabilities of traditional architectures where many functionalities do not exist in distinct modular entities. This paper reports on an investigation of how and where some distinct design requirements lead to crosscut-ting concerns when decomposed into code in goal models such as KAOS. We begin by matching our past experience in aspect discovery at the code level with a detailed require-ments modeling of the same architecture in KAOS. The dis-covered patterns are validated in an independent project where the requirements modeling and the aspect identifi-cation are separately conducted. We observe that satisfy-ing OR-decomposed subgoals in the KAOS model typically leads to tangled implementations, and agents responsible for multiple OR-refined goals should be implemented in the aspect-oriented manner." ++++ diff --git a/content/publications/load-balancing-content-based-publish-subscribe-systems.md b/content/publications/load-balancing-content-based-publish-subscribe-systems.md new file mode 100644 index 0000000..5b695ef --- /dev/null +++ b/content/publications/load-balancing-content-based-publish-subscribe-systems.md @@ -0,0 +1,10 @@ ++++ +title = "Load Balancing Content-Based Publish/Subscribe Systems" +year = 2010 +authors = ["Alex King Yeung Cheung", "Hans-Arno Jacobsen"] +venue = "ACM Transactions on Computer Systems" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.1145/1880018.1880020" +abstract = "Distributed content-based publish/subscribe systems suffer from performance degradation and poor scalability caused by uneven load distributions typical in real-world applications. The reason for this shortcoming is the lack of a load balancing scheme. This article proposes a load balancing solution specifically tailored to the needs of content-based publish/subscribe systems that is distributed, dynamic, adaptive, transparent, and accommodates heterogeneity. The solution consists of three key contributions: a load balancing framework, a novel load estimation algorithm, and three offload strategies. A working prototype of our solution is built on an open-sourced content-based publish/subscribe system and evaluated on PlanetLab, a cluster testbed, and in simulations. Real-life experiment results show that the proposed load balancing solution is efficient with less than 0.2% overhead; effective in distributing and balancing load originating from a single server to all available servers in the network; and capable of preventing overloads to preserve system stability, availability, and quality of service." ++++ diff --git a/content/publications/logstore-a-workload-aware-adaptable-key-value-store-on-hybrid-storage-systems.md b/content/publications/logstore-a-workload-aware-adaptable-key-value-store-on-hybrid-storage-systems.md new file mode 100644 index 0000000..3cf0da1 --- /dev/null +++ b/content/publications/logstore-a-workload-aware-adaptable-key-value-store-on-hybrid-storage-systems.md @@ -0,0 +1,10 @@ ++++ +title = "LogStore: A Workload-Aware, Adaptable Key-Value Store on Hybrid Storage Systems" +year = 2022 +authors = ["Prashanth Menon", "Thamir M. Qadah", "Tilmann Rabl", "Mohammad Sadoghi", "Hans-Arno Jacobsen"] +venue = "IEEE Transactions on Knowledge and Data Engineering" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.1109/tkde.2020.3027191" +abstract = "Due to recent explosion of data volume and velocity, a new array of lightweight key-value stores have emerged to serve as alternatives to traditional databases. The majority of these storage engines, however, sacrifice their read performance in order to cope with write throughput by avoiding random disk access when writing a record in favor of fast sequential accesses. But, the boundary between sequential versus random access is becoming blurred with the advent of solid-state drives (SSDs). In this work, we propose our new key-value store, LogStore, optimized for hybrid storage architectures. Additionally, introduce a novel cost-based data staging model based on log-structured storage, in which recent changes are first stored on SSDs, and pushed to HDD as it ages, while minimizing the read/write amplification for merging data from SSDs and HDDs. Furthermore, we take a holistic approach in improving both the read and write performance by dynamically optimizing the data layout, such as deferring and reversing the compaction process, and developing an access strategy to leverage the strengths of each available medium in our storage hierarchy. Lastly, in our extensive evaluation, we demonstrate that LogStore achieves up to 6x improvement in throughput/latency over LevelDB, a state-of-the-art key-value store." ++++ diff --git a/content/publications/making-crdts-not-so-eventual.md b/content/publications/making-crdts-not-so-eventual.md new file mode 100644 index 0000000..2a0fab8 --- /dev/null +++ b/content/publications/making-crdts-not-so-eventual.md @@ -0,0 +1,10 @@ ++++ +title = "Making CRDTs Not So Eventual" +year = 2024 +authors = ["Yunhao Mao", "Gengrui Zhang", "Zongxin Liu", "Pezhman Nasirifard", "Sofia Tijanic", "Hans-Arno Jacobsen"] +venue = "Proceedings of the VLDB Endowment" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.14778/3705829.3705850" +abstract = "Conflict-free replicated data types (CRDTs) are highly available and performant data replication solutions for distributed applications. However, their eventual consistency guarantees are often insufficient for ensuring application correctness, especially in the presence of Byzantine failures. Naively applying traditional consensus and Byzantine fault tolerance (BFT) protocols to CRDT updates for stronger guarantees, while intuitive, negates the performance benefits of CRDTs. We introduce a novel programming model called reliable CRDTs that expands CRDTs with additional guarantees: users can query strongly or eventually consistent values, enforce a total order among selected operations, and define data-type level invariants while remaining operational in the presence of Byzantine failures. Reliable CRDTs enable the use of CRDTs in scenarios where strong consistency is needed while maintaining their performance advantages. We present an implementation of reliable CRDTs named Janus. It enhances CRDTs with the aforementioned features by functioning as a middleware that facilitates CRDT communication and asynchronously runs a BFT consensus protocol. Our evaluation demonstrates that Janus achieves 21× higher throughput than naively applying state-of-the-art BFT protocols such as HotStuff achieves, and it remains responsive even under heavy loads." ++++ diff --git a/content/publications/mar-fl-a-communication-efficient-peer-to-peer-federated-learning-system.md b/content/publications/mar-fl-a-communication-efficient-peer-to-peer-federated-learning-system.md new file mode 100644 index 0000000..e2f8cd9 --- /dev/null +++ b/content/publications/mar-fl-a-communication-efficient-peer-to-peer-federated-learning-system.md @@ -0,0 +1,10 @@ ++++ +title = "MAR-FL: A Communication Efficient Peer-to-Peer Federated Learning System" +year = 2025 +authors = ["Felix Mulitze", "Herbert Woisetschläger", "Hans-Arno Jacobsen"] +venue = "AI4NextG Workshop at NeurIPS 2025" +publication_type = "Workshop Paper" +research = ["distributed-machine-learning"] +external_url = "https://neurips.cc/virtual/2025/123210" +abstract = "The convergence of next-generation wireless systems and distributed Machine Learning (ML) demands Federated Learning (FL) methods that remain efficient and robust with wireless connected peers and under network churn. Peer-to-peer (P2P) FL removes the bottleneck of a central coordinator, but existing approaches suffer from excessive communication complexity, limiting their scalability in practice. We introduce MAR-FL, a novel P2P FL system that leverages iterative group-based aggregation to substantially reduce communication overhead while retaining resilience to churn. MAR-FL achieves communication costs that scale as O(N log N), contrasting with the O(N^2) complexity of previously existing baselines, and thereby maintains effectiveness especially as the number of peers in an aggregation round grows. The system is robust towards unreliable FL clients and can integrate private computing." ++++ diff --git a/content/publications/materialized-views-in-cassandra.md b/content/publications/materialized-views-in-cassandra.md new file mode 100644 index 0000000..924b7ec --- /dev/null +++ b/content/publications/materialized-views-in-cassandra.md @@ -0,0 +1,10 @@ ++++ +title = "Materialized views in Cassandra" +year = 2014 +authors = ["Tilmann Rabl", "Hans-Arno Jacobsen"] +venue = "CASCON" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://dl.acm.org/citation.cfm?id=2735581" +abstract = "Many web companies deal with enormous data sizes and request rates beyond the capabilities of traditional database systems. This has led to the de-velopment of modern Big Data Platforms (BDPs). BDPs handle large amounts of data and activity through massively distributed infrastructures. To achieve performance and availability at Internet scale, BDPs restrict querying capability, and pro-vide weaker consistency guarantees than traditional ACID transactions. The reduced functionality as found in key-value stores is sufficient for many web applications. An important requirement of many big data sys-tems is an online view of the current status of the data and activity. Typical big data systems such as key-value stores only allow a key-based access. In order to enable more complex querying mecha-nisms, while satisfying necessary latencies materi-alized views are employed. The efficiency of the maintenance of these views is a key factor of the usability of the system. Expensive operations such as full table scans are impractical for small, fre-quent modifications on Internet-scale data sets. In this paper, we present an efficient implementation of materialized views in key-value stores that en-ables complex query processing and is tailored for efficient maintenance." ++++ diff --git a/content/publications/mdms-music-data-matching-system-for-query-variant-retrieval.md b/content/publications/mdms-music-data-matching-system-for-query-variant-retrieval.md new file mode 100644 index 0000000..d914f15 --- /dev/null +++ b/content/publications/mdms-music-data-matching-system-for-query-variant-retrieval.md @@ -0,0 +1,10 @@ ++++ +title = "MDMS: Music Data Matching System for Query Variant Retrieval" +year = 2021 +authors = ["Rinita Roy", "Ruben Mayer", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 29th ACM International Conference on Multimedia" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/3474085.3478551" +abstract = "The distribution of royalty fees to music right holders is slow and inefficient due to the lack of automation in music recognition and music licensing processes. The challenge for an improved system is to recognise different versions of a music such as remix or cover versions, leading to clear assessment and unique identification of each music work. Through our music data matching system called MDMS, we query many indexed and stored music pieces with a small part of a music piece. The system retrieves the closest stored variant of the input query by using music fingerprints of the underlying melody together with signal processing techniques. Tailored indices based on fingerprint hashes accelerate processing across a large corpus of stored music. Results are found even if the stored versions vary from the query song in terms of one or more music features --- tempo, key/mode, presence of instruments/vocals, and singer --- and the differences are highlighted in the output." ++++ diff --git a/content/publications/measurement-system-and-dataset-for-in-depth-analysis-of-appliance-energy-consumption-in-industrial-environment.md b/content/publications/measurement-system-and-dataset-for-in-depth-analysis-of-appliance-energy-consumption-in-industrial-environment.md new file mode 100644 index 0000000..c3de59d --- /dev/null +++ b/content/publications/measurement-system-and-dataset-for-in-depth-analysis-of-appliance-energy-consumption-in-industrial-environment.md @@ -0,0 +1,10 @@ ++++ +title = "Measurement system and dataset for in-depth analysis of appliance energy consumption in industrial environment" +year = 2019 +authors = ["Matthias Kahl", "Veronika Krause", "Rudolph Hackenberg", "Anwar Ul Haq", "Anton Horn", "Hans-Arno Jacobsen", "Thomas Kriechbaumer", "Michael Petzenhauser", "Mikhail Shamonin", "Anton Udalzow"] +venue = "tm - Technisches Messen" +publication_type = "Journal Article" +research = [] +external_url = "https://doi.org/10.1515/teme-2018-0038" +abstract = "To support a rational and efficient use of electrical energy in residential and industrial environments, Non-Intrusive Load Monitoring (NILM) provides several techniques to identify state and power consumption profiles of connected appliances. Design requirements for such systems include a low hardware and installations costs for residential, reliability and high-availability for industrial purposes, while keeping invasive interventions into the electrical infrastructure to a minimum. This work introduces a reference hardware setup that allows an in depth analysis of electrical energy consumption in industrial environments. To identify appliances and their consumption profile, appropriate identification algorithms are developed by the NILM community. To enable an evaluation of these algorithms on industrial appliances, we introduce the Laboratory-measured IndustriaL Appliance Characteristics (LILAC) dataset: 1302 measurements from one, two, and three concurrently running appliances of 15 appliance types, measured with the introduced testbed. To allow in-depth appliance consumption analysis, measurements were carried out with a sampling rate of 50 kHz and 16-bit amplitude resolution for voltage and current signals. We show in experiments that signal signatures, contained in the measurement data, allows one to distinguish the single measured electrical appliances with a baseline machine learning approach of nearly 100 % accuracy." ++++ diff --git a/content/publications/medal-a-cost-effective-high-frequency-energy-data-acquisition-system-for-electrical-appliances.md b/content/publications/medal-a-cost-effective-high-frequency-energy-data-acquisition-system-for-electrical-appliances.md new file mode 100644 index 0000000..9010366 --- /dev/null +++ b/content/publications/medal-a-cost-effective-high-frequency-energy-data-acquisition-system-for-electrical-appliances.md @@ -0,0 +1,10 @@ ++++ +title = "MEDAL: A Cost-Effective High-Frequency Energy Data Acquisition System for Electrical Appliances" +year = 2017 +authors = ["Thomas Kriechbaumer", "Anwar Ul Haq", "Matthias Kahl", "Hans-Arno Jacobsen"] +venue = "Proceedings of the Eighth International Conference on Future Energy Systems" +publication_type = "Conference Paper" +research = [] +external_url = "https://doi.org/10.1145/3077839.3077844" +abstract = "Traditional energy measurement fails to provide support to consumers to make intelligent decisions to save energy. Non-intrusive load monitoring is one solution that provides disaggregated power consumption profiles. Machine learning approaches rely on public datasets to train parameters for their algorithms, most of which only provide low-frequency appliance-level measurements, thus limiting the available feature space for recognition." ++++ diff --git a/content/publications/meed-an-unsupervised-multi-environment-event-detector-for-non-intrusive-load-monitoring.md b/content/publications/meed-an-unsupervised-multi-environment-event-detector-for-non-intrusive-load-monitoring.md new file mode 100644 index 0000000..f42ba11 --- /dev/null +++ b/content/publications/meed-an-unsupervised-multi-environment-event-detector-for-non-intrusive-load-monitoring.md @@ -0,0 +1,10 @@ ++++ +title = "MEED: An Unsupervised Multi-Environment Event Detector for Non-Intrusive Load Monitoring" +year = 2019 +authors = ["Daniel Jorde", "Matthias Kahl", "Hans-Arno Jacobsen"] +venue = "2019 IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids (SmartGridComm)" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/smartgridcomm.2019.8909729" +abstract = "The accurate detection of transitions between appliance states in electrical signals is the fundamental step that numerous energy conserving applications, such as Non-Intrusive Load Monitoring, rely on. So far, domain experts define rules and patterns to detect changes of appliance states and to extract detailed consumption information of individual appliances subsequently. Such event detectors are specifically designed for certain environments and need to be tediously adapted for new ones, as they require in-depth expert knowledge of the environment. To overcome this limitation, we propose a new unsupervised, multi-environment event detector, called MEED, that is based on a bidirectional recurrent denoising autoencoder. The performance of MEED is evaluated by comparing it to two state-of-the-art algorithms on two publicly available datasets from different environments. The results show that MEED improves the current state of the art and outperforms the reference algorithms on a residential (BLUED) and an office environment (BLOND) dataset while being trained and used fully unsupervised in the heterogeneous environments." ++++ diff --git a/content/publications/merc-match-at-edge-and-route-intra-cluster-for-content-based-publish-subscribe-systems.md b/content/publications/merc-match-at-edge-and-route-intra-cluster-for-content-based-publish-subscribe-systems.md new file mode 100644 index 0000000..fed0885 --- /dev/null +++ b/content/publications/merc-match-at-edge-and-route-intra-cluster-for-content-based-publish-subscribe-systems.md @@ -0,0 +1,10 @@ ++++ +title = "MERC: Match at Edge and Route intra-Cluster for Content-based Publish/Subscribe Systems" +year = 2015 +authors = ["Shuping Ji", "Chunyang Ye", "Jun Wei", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 16th Annual Middleware Conference" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/2814576.2814801" +abstract = "Despite suffering from inefficiency and flexibility limitations, the filter-based routing (FBR) algorithm is widely used in content-based publish/subscribe (pub/sub) systems. To address its limitations, we propose a dynamic destination-based routing algorithm called D-DBR, which decomposes pub/sub into two independent parts: Content-based matching and destination-based multicasting. D-DBR exhibits low event matching cost and high efficiency, flexibility, and robustness for event routing in small scale overlays. To boost scalability, we further complement D-DBR with a new routing algorithm called MERC. MERC divides the overlay into interconnected clusters and applies content-based and destination-based mechanisms to route events inter- and intra-cluster, respectively. We implemented all algorithms in the PADRES pub/sub system. Experimental results show that our algorithms outperform FBR in terms of improving event dissemination throughput by up to 700% and reducing the end-to-end latency by up to 55%." ++++ diff --git a/content/publications/mess-dynamically-learned-inference-time-llm-routing-in-model-zoos-with-service-level-guarantees.md b/content/publications/mess-dynamically-learned-inference-time-llm-routing-in-model-zoos-with-service-level-guarantees.md new file mode 100644 index 0000000..c26ff6f --- /dev/null +++ b/content/publications/mess-dynamically-learned-inference-time-llm-routing-in-model-zoos-with-service-level-guarantees.md @@ -0,0 +1,10 @@ ++++ +title = "MESS+: Dynamically Learned Inference-Time LLM Routing in Model Zoos with Service Level Guarantees" +year = 2025 +authors = ["Herbert Woisetschläger", "Ryan Zhang", "Shiqiang Wang", "Hans-Arno Jacobsen"] +venue = "Advances in Neural Information Processing Systems 38" +publication_type = "Conference Paper" +research = ["distributed-machine-learning"] +external_url = "https://doi.org/10.52202/085713-1804" +abstract = "Open-weight large language model (LLM) zoos provide access to numerous high-quality models, but selecting the appropriate model for specific tasks remains challenging and requires technical expertise. Most users simply want factually correct, safe, and satisfying responses without concerning themselves with model technicalities, while inference service providers prioritize minimizing operating costs. These competing interests are typically mediated through service level agreements (SLAs) that guarantee minimum service quality. We introduce MESS+, a stochastic optimization algorithm for cost-optimal LLM request routing while providing rigorous SLA compliance guarantees. MESS+ learns request satisfaction probabilities of LLMs in real-time as users interact with the system, based on which model selection decisions are made by solving a per-request optimization problem. Our algorithm includes a novel combination of virtual queues and request satisfaction prediction, along with a theoretical analysis of cost optimality and constraint satisfaction. Across a wide range of state-of-the-art LLM benchmarks, MESS+ achieves an average of $2\\times$ cost savings compared to existing LLM routing techniques." ++++ diff --git a/content/publications/mess-energy-optimal-inferencing-in-language-model-zoos-with-service-level-guarantees.md b/content/publications/mess-energy-optimal-inferencing-in-language-model-zoos-with-service-level-guarantees.md new file mode 100644 index 0000000..4728115 --- /dev/null +++ b/content/publications/mess-energy-optimal-inferencing-in-language-model-zoos-with-service-level-guarantees.md @@ -0,0 +1,10 @@ ++++ +title = "MESS+: Energy-Optimal Inferencing in Language Model Zoos with Service Level Guarantees" +year = 2024 +authors = ["Ryan Zhang", "Herbert Woisetschläger", "Shiqiang Wang", "Hans-Arno Jacobsen"] +venue = "Workshop on Adaptive Foundation Models at NeurIPS 2024" +publication_type = "Workshop Paper" +research = ["distributed-machine-learning"] +external_url = "https://openreview.net/forum?id=OoReeQpwmW" +abstract = "Open-weight large language model (LLM) zoos allow users to quickly integrate state-of-the-art models into systems. Despite increasing availability, selecting the most appropriate model for a given task still largely relies on public benchmark leaderboards and educated guesses. This can be unsatisfactory for both inference service providers and end users, where the providers usually prioritize cost efficiency, while the end users usually prioritize model output quality for their inference requests. In commercial settings, these two priorities are often brought together in Service Level Agreements (SLA). We present MESS+, an online stochastic optimization algorithm for energy-optimal model selection from a model zoo, which works on a per-inference-request basis. For a given SLA that requires high accuracy, we are up to 2.5x more energy efficient with MESS+ than with randomly selecting an LLM from the zoo while maintaining SLA quality constraints." ++++ diff --git a/content/publications/methods-for-quantifying-energy-consumption-in-tpc-h.md b/content/publications/methods-for-quantifying-energy-consumption-in-tpc-h.md new file mode 100644 index 0000000..c21f45d --- /dev/null +++ b/content/publications/methods-for-quantifying-energy-consumption-in-tpc-h.md @@ -0,0 +1,10 @@ ++++ +title = "Methods for Quantifying Energy Consumption in TPC-H" +year = 2018 +authors = ["Meikel Poess", "Da Qi Ren", "Tilmann Rabl", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 2018 ACM/SPEC International Conference on Performance Engineering" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/3184407.3184429" +abstract = "Historically, performance and price-performance of computer systems have been the key purchasing arguments for customers. However, with rising energy costs and increasing power consumption due to the ever-growing demand for compute power (servers, storage, networks), electricity bills have become a significant expense for today»s data centers. In order to measure energy consumption in standardized ways, the Standard Performance Evaluation Corporation (SPEC) has developed a benchmark dedicated to measuring the power consumption of single servers (SPECpower\\_ssj2008), while the Transaction Processing Performance Council (TPC) and the Storage Performance Council (SPC) have developed general specifications that govern how energy is measured for any of its benchmarks. Energy reporting is optional in TPC and SPC results. While there are close to 600 SPECpower\\_ssj2008 results, there have been only three TPC and no SPC benchmark results published that report energy consumption. In this paper, we argue that the low number of TPC publications is due to the large setups required in TPC benchmarks and the, subsequently, complicated measurement setup. Running on a typical big data setup we evaluate two alternative methods to quantify energy consumption during TPC-H's multi-user runs, namely by taking measurements of on-chip power sensors controlled through Intelligent Platform Management Interface and by estimating power consumption via the nameplate power consumption method. We compare these later two methods with power measurements taken from external power meters as required by SPEC and TPC benchmarks." ++++ diff --git a/content/publications/minimal-broker-overlay-design-for-content-based-publish-subscribe-systems.md b/content/publications/minimal-broker-overlay-design-for-content-based-publish-subscribe-systems.md new file mode 100644 index 0000000..83f9a7a --- /dev/null +++ b/content/publications/minimal-broker-overlay-design-for-content-based-publish-subscribe-systems.md @@ -0,0 +1,10 @@ ++++ +title = "Minimal broker overlay design for content-based publish/subscribe systems" +year = 2013 +authors = ["Naweed Tajuddin", "Balasubramaneyam Maniymaran", "Hans-Arno Jacobsen"] +venue = "CASCON" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://dl.acm.org/citation.cfm?id=2555528" +abstract = "Mission-critical distributed applications, such as Internet advertising platforms, increasingly utilize distributed publish/subscribe systems as a messag-ing substrate for information dissemination. These applications require low latency performance from the substrate, as the timely delivery of messages can have a direct impact on revenue. The cost of managing and operating distributed publish/sub-scribe systems, however, can be prohibitive due to system size and scale. It is, therefore, critical to de-rive low latency message delivery from a minimal set of system resources. To this end, this paper presents a solution for designing low latency, minimal-broker overlay networks for content-based publish/subscribe sys-tems. The solution is developed in two parts. First, a framework is developed to quantify the similarity of entities in content-based publish/subscribe sys-tems. Second, algorithms are presented for design-ing overlays that utilize a minimal number of bro-kers in order to provide low latency performance at reduced cost." ++++ diff --git a/content/publications/minimizing-the-communication-cost-of-aggregation-in-publish-subscribe-systems.md b/content/publications/minimizing-the-communication-cost-of-aggregation-in-publish-subscribe-systems.md new file mode 100644 index 0000000..a0ae11e --- /dev/null +++ b/content/publications/minimizing-the-communication-cost-of-aggregation-in-publish-subscribe-systems.md @@ -0,0 +1,10 @@ ++++ +title = "Minimizing the Communication Cost of Aggregation in Publish/Subscribe Systems" +year = 2015 +authors = ["Navneet Kumar Pandey", "Kaiwen Zhang", "Stéphane Weiss", "Hans-Arno Jacobsen", "Roman Vitenberg"] +venue = "2015 IEEE 35th International Conference on Distributed Computing Systems" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icdcs.2015.54" +abstract = "Modern applications for distributed publish/subscribe systems often require stream aggregation capabilities along with rich data filtering. When compared to other distributed systems, aggregation in pub/sub differentiates itself as a complex problem which involves dynamic dissemination paths that are difficult to predict and optimize for a priori, temporal fluctuations in publication rates, and the mixed presence of aggregated and non-aggregated workloads. In this paper, we propose a formalization for the problem of minimizing communication traffic in the context of aggregation in pub/sub. We present a solution to this minimization problem by using a reduction to the well-known problem of minimum vertex cover in a bipartite graph. This solution is optimal under the strong assumption of complete knowledge of future publications. We call the resulting algorithm \"Aggregation Decision, Optimal with Complete Knowledge\" (ADOCK). We also show that under a dynamic setting without full knowledge, ADOCK can still be applied to produce a low, yet not necessarily optimal, communication cost. We also devise a computationally cheaper dynamic approach called \"Aggregation Decision with Weighted Publication\" (WAD). We compare our solutions experimentally using two real datasets and explore the trade-offs with respect to communication and computation costs." ++++ diff --git a/content/publications/minimum-delay-multicast-algorithms-for-mesh-overlays.md b/content/publications/minimum-delay-multicast-algorithms-for-mesh-overlays.md new file mode 100644 index 0000000..3b54fdb --- /dev/null +++ b/content/publications/minimum-delay-multicast-algorithms-for-mesh-overlays.md @@ -0,0 +1,10 @@ ++++ +title = "Minimum-Delay Multicast Algorithms for Mesh Overlays" +year = 2015 +authors = ["Kianoosh Mokhtarian", "Hans-Arno Jacobsen"] +venue = "IEEE/ACM Transactions on Networking" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.1109/tnet.2014.2310735" +abstract = "We study delivering delay-sensitive data to a group of receivers with minimum latency. This latency consists of the time that the data spends in overlay links as well as the delay incurred at each overlay node, which has to send out a piece of data several times over a finite-capacity network connection. The latter part is a significant portion of the total delay as we show in the paper, yet it is often ignored or only partially addressed by previous multicast algorithms. We analyze the actual delay in multicast trees and consider building trees with minimum-average and minimum-maximum delay. We show the NP-hardness of these problems and prove that they cannot be approximated in polynomial time to within any reasonable approximation ratio. We then present a set of algorithms to build minimum-delay multicast trees that cover a wide range of application requirements-min-average and min-max delay, for different scales, real-time requirements, and session characteristics. We conduct comprehensive experiments on different real-world datasets, using various overlay network models. The results confirm that our algorithms can achieve much lower delays (up to 60% less) and up to orders-of-magnitude faster running times (i.e., supporting larger scales) than previous related approaches." ++++ diff --git a/content/publications/minimum-delay-overlay-multicast.md b/content/publications/minimum-delay-overlay-multicast.md new file mode 100644 index 0000000..5238ca1 --- /dev/null +++ b/content/publications/minimum-delay-overlay-multicast.md @@ -0,0 +1,10 @@ ++++ +title = "Minimum-delay overlay multicast" +year = 2013 +authors = ["Kianoosh Mokhtarian", "Hans-Arno Jacobsen"] +venue = "2013 Proceedings IEEE INFOCOM" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/infcom.2013.6566975" +abstract = "Delivering delay-sensitive data to a group of receivers with minimum latency is a fundamental problem for various distributed applications. In this paper, we study multicast routing with minimum end-to-end delay to the receivers. The delay to each receiver in a multicast tree consist of the time that the data spends in overlay links as well as the latency incurred at each overlay node, which has to send out a piece of data several times over a finite-capacity network connection. The latter portion of the delay, which is proportional to the degree of nodes in the tree, can be a significant portion of the total delay as we show in the paper. Yet, it is often ignored or only partially addressed by previous multicast algorithms. We formulate the actual delay to the receivers in a multicast tree and consider minimizing the average and the maximum delay in the tree. We show the NP-hardness of these problems and prove that they cannot be approximated in polynomial time to within any reasonable approximation ratio. We then present a number of efficient algorithms to build a multicast tree in which the average or the maximum delay is minimized. These algorithms cover a wide range of overlay sizes for both versions of our problem. The effectiveness of our algorithms is demonstrated through comprehensive experiments on different real-world datasets, and using various overlay network models. The results confirm that our algorithms can achieve much lower delays (up to 60% less) and up to orders of magnitude faster running times (i.e., supporting larger scales) than previous minimum-delay multicast approaches." ++++ diff --git a/content/publications/mining-crosscutting-concerns-through-random-walks.md b/content/publications/mining-crosscutting-concerns-through-random-walks.md new file mode 100644 index 0000000..1406474 --- /dev/null +++ b/content/publications/mining-crosscutting-concerns-through-random-walks.md @@ -0,0 +1,10 @@ ++++ +title = "Mining Crosscutting Concerns through Random Walks" +year = 2012 +authors = ["Charles Zhang", "Hans-Arno Jacobsen"] +venue = "IEEE Transactions on Software Engineering" +publication_type = "Journal Article" +research = [] +external_url = "https://doi.org/10.1109/tse.2011.83" +abstract = "Inspired by our past manual aspect mining experiences, this paper describes a probabilistic random walk model to approximate the process of discovering crosscutting concerns (CCs) in the absence of the domain knowledge about the investigated application. The random walks are performed on the concept graphs extracted from the program sources to calculate metrics of “utilization” and “aggregation” for each of the program elements. We rank all the program elements based on these metrics and use a threshold to produce a set of candidates that represent crosscutting concerns. We implemented the algorithm as the Prism CC miner (PCM) and evaluated PCM on Java applications ranging from a small-scale drawing application to a medium-sized middleware application and to a large-scale enterprise application server. Our quantification shows that PCM is able to produce comparable results (95 percent accuracy for the top 125 candidates) with respect to the manual mining effort. PCM is also significantly more effective as compared to the conventional approach." ++++ diff --git a/content/publications/mocha-scalable-and-compliant-function-scheduling-for-federated-serverless-computing.md b/content/publications/mocha-scalable-and-compliant-function-scheduling-for-federated-serverless-computing.md new file mode 100644 index 0000000..626417f --- /dev/null +++ b/content/publications/mocha-scalable-and-compliant-function-scheduling-for-federated-serverless-computing.md @@ -0,0 +1,10 @@ ++++ +title = "Mocha: Scalable and Compliant Function Scheduling for Federated Serverless Computing" +year = 2025 +authors = ["Yuqiu Zhang", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 26th International Middleware Conference" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/3721462.3770780" +abstract = "Serverless computing promises on-demand elasticity and simplified deployment, yet today's production-grade serverless platforms remain tied to a single-provider, centrally scheduled control plane. This centralized scheduling model faces mounting challenges in handling heterogeneous policies, data governance constraints, and dynamic workloads for the modern web, where applications increasingly span multiple geo-distributed autonomous administrative domains. In this paper, we present Mocha, a decentralized, policy-aware framework for scheduling serverless functions across a federated ecosystem. At its core, Mocha proposes a hierarchically structured distributed hash table that embeds geographical and organizational context to facilitate locality-aware scheduling without any central authority. By implementing a formally specified compliance engine at each domain, Mocha guarantees that all regulatory, locality, and resource constraints are honored for function placement decisions. Experiments show that Mocha reduces scheduling tail latency by 4–9× compared to alternatives while maintaining full policy adherence." ++++ diff --git a/content/publications/model-based-dispatch-strategies-for-lithium-ion-battery-energy-storage-applied-to-pay-as-bid-markets-for-secondary-reserve.md b/content/publications/model-based-dispatch-strategies-for-lithium-ion-battery-energy-storage-applied-to-pay-as-bid-markets-for-secondary-reserve.md new file mode 100644 index 0000000..69cc0ce --- /dev/null +++ b/content/publications/model-based-dispatch-strategies-for-lithium-ion-battery-energy-storage-applied-to-pay-as-bid-markets-for-secondary-reserve.md @@ -0,0 +1,10 @@ ++++ +title = "Model-Based Dispatch Strategies for Lithium-Ion Battery Energy Storage Applied to Pay-as-Bid Markets for Secondary Reserve" +year = 2017 +authors = ["Christoph Goebel", "Holger Hesse", "Michael Schimpe", "Andreas Jossen", "Hans-Arno Jacobsen"] +venue = "IEEE Transactions on Power Systems" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.1109/tpwrs.2016.2626392" +abstract = "Due to their decreasing cost, lithium-ion batteries (LiB) are becoming increasingly attractive for grid-scale applications. In this paper, we investigate the use of LiB for providing secondary reserve and show how the achieved cost savings could be increased by using model-based optimization techniques. In particular, we compare a maximum use dispatch strategy with two different cost-minimizing strategies. For the estimation of state-dependent battery usage cost, we combine an existing electro-thermal LiB model of a mature lithium-iron-phosphate battery cell with corresponding semiempirical calendar and cycle aging models. We estimate the benefit of storage operation from the system operator's point of view by gauging the avoided cost of activated reserve. Our evaluation is based on two years worth of data from the German reserve market. The proposed cost minimizing dispatch strategies yield significantly better results than a dispatch strategy that maximizes battery utilization." ++++ diff --git a/content/publications/modeling-uncertainties-in-publish-subscribe-systems.md b/content/publications/modeling-uncertainties-in-publish-subscribe-systems.md new file mode 100644 index 0000000..792bbc8 --- /dev/null +++ b/content/publications/modeling-uncertainties-in-publish-subscribe-systems.md @@ -0,0 +1,10 @@ ++++ +title = "Modeling Uncertainties in Publish/Subscribe Systems" +year = 2004 +authors = ["Haifeng Liu", "Hans-Arno Jacobsen"] +venue = "Proceedings. 20th International Conference on Data Engineering" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icde.2004.1320023" +abstract = "In the publish/subscribe paradigm, information providers disseminate publications to all consumers who have expressed interest by registering subscriptions. This paradigm has found wide-spread applications, ranging from selective information dissemination to network management. However, all existing publish/subscribe systems cannot capture uncertainty inherent to the information in either subscriptions or publications. In many situations, exact knowledge of either specific subscriptions or publications is not available. Moreover, especially in selective information dissemination applications, it is often more appropriate for a user to formulate her search requests or information offers in less precise terms, rather than defining a sharp limit. To address these problems, this paper proposes a new publish/subscribe model based on possibility theory and fuzzy set theory to process uncertainties for both subscriptions and publications. Furthermore, an approximate publish/subscribe matching problem is defined and algorithms for solving it are developed and evaluated." ++++ diff --git a/content/publications/mothpad-monitoring-pub-sub-activity-in-cyber-physical-systems.md b/content/publications/mothpad-monitoring-pub-sub-activity-in-cyber-physical-systems.md new file mode 100644 index 0000000..4c3d39f --- /dev/null +++ b/content/publications/mothpad-monitoring-pub-sub-activity-in-cyber-physical-systems.md @@ -0,0 +1,10 @@ ++++ +title = "MothPad: monitoring pub/sub activity in cyber-physical systems" +year = 2016 +authors = ["César Cañas", "Kaiwen Zhang", "Bettina Kemme", "Jörg Kienzle", "Hans-Arno Jacobsen"] +venue = "CASCON" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://dl.acm.org/citation.cfm?id=3049898" +abstract = "Content-based publish/subscribe is an attractive option for disseminating event data in cyber-physical systems. To this end, we propose MothPad: a monitoring and visualization tool to demonstrate the performance of various pub/sub solutions within the context of location-based applications. MothPad consists of Mammoth, an online game research framework used as a cyber-physical system simulator, and PADRES, the publish/subscribe dissemination substrate. Both are instrumented and the performance is displayed in real-time using a monitoring client. We show the applicability of our approach through two case studies: network engines for online games and self-evolving subscriptions." ++++ diff --git a/content/publications/multi-client-transactions-in-distributed-publish-subscribe-systems.md b/content/publications/multi-client-transactions-in-distributed-publish-subscribe-systems.md new file mode 100644 index 0000000..9722b79 --- /dev/null +++ b/content/publications/multi-client-transactions-in-distributed-publish-subscribe-systems.md @@ -0,0 +1,10 @@ ++++ +title = "Multi-Client Transactions in Distributed Publish/Subscribe Systems" +year = 2018 +authors = ["Martin Jergler", "Kaiwen Zhang", "Hans-Arno Jacobsen"] +venue = "2018 IEEE 38th International Conference on Distributed Computing Systems (ICDCS)" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icdcs.2018.00022" +abstract = "Transactional operation processing among clients is increasingly required of publish/subscribe (pub/sub) systems in enterprise settings. For instance, in workflow management, dispatching or consolidating process instances require publications and (un-) subscriptions by different clients to be executed according to ACID semantics. As pub/sub systems are usually optimized for performance and scalability, such properties are often neglected, which results in unexpected system behavior. In this paper, we provide a model for supporting multiclient transactions in pub/sub. We formalize ACID properties for pub/sub, and define a consistency model and isolation level required in the aforementioned scenarios. We present three approaches for two transaction types: S-TX, where a coordinator has full static knowledge about all operations in a transaction, and D-TX/D-TXNI, where operations by other clients are dynamic and unknown to the coordinator. We describe algorithms realizing these approaches and experimentally evaluate them by comparing to a baseline mechanism, which simulates these guarantees partially with manual waits between operations. Our results show that the uncertainty introduced by the dynamic behavior renders D-TX/D-TXNI costly, and suitable only for small configurations or rare occasions. S-TX, in contrast, offers enriched semantics for many applications in a scalable manner without disrupting regular event routing." ++++ diff --git a/content/publications/multi-query-stream-processing-on-fpgas.md b/content/publications/multi-query-stream-processing-on-fpgas.md new file mode 100644 index 0000000..fa46113 --- /dev/null +++ b/content/publications/multi-query-stream-processing-on-fpgas.md @@ -0,0 +1,10 @@ ++++ +title = "Multi-query Stream Processing on FPGAs" +year = 2012 +authors = ["Mohammad Sadoghi", "Rija Javed", "Naif Tarafdar", "Harsh Singh", "Rohan Palaniappan", "Hans-Arno Jacobsen"] +venue = "2012 IEEE 28th International Conference on Data Engineering" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icde.2012.39" +abstract = "We present an efficient multi-query event stream platform to support query processing over high-frequency event streams. Our platform is built over reconfigurable hardware -- FPGAs -- to achieve line-rate multi-query processing by exploiting unprecedented degrees of parallelism and potential for pipelining, only available through custom-built, application-specific and low-level logic design. Moreover, a multi-query event stream processing engine is at the core of a wide range of applications including real-time data analytics, algorithmic trading, targeted advertisement, and (complex) event processing." ++++ diff --git a/content/publications/neurondepot-keeping-your-colleagues-in-sync-by-combining-modern-cloud-storage-services-the-local-file-system-and-simple-web-applications.md b/content/publications/neurondepot-keeping-your-colleagues-in-sync-by-combining-modern-cloud-storage-services-the-local-file-system-and-simple-web-applications.md new file mode 100644 index 0000000..79055dc --- /dev/null +++ b/content/publications/neurondepot-keeping-your-colleagues-in-sync-by-combining-modern-cloud-storage-services-the-local-file-system-and-simple-web-applications.md @@ -0,0 +1,10 @@ ++++ +title = "NeuronDepot: keeping your colleagues in sync by combining modern cloud storage services, the local file system, and simple web applications" +year = 2014 +authors = ["Philipp L. Rautenberg", "Ajayrama Kumaraswamy", "Álvaro Tejero-Cantero", "Christoph Doblander", "Mohammad Norouzian", "Kazuki Kai", "Hans-Arno Jacobsen", "Hiroyuki Ai", "Thomas Wachtler", "Hidetoshi Ikeno"] +venue = "Frontiers in Neuroinformatics" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.3389/fninf.2014.00055" +abstract = "Neuroscience today deals with a \"data deluge\" derived from the availability of high-throughput sensors of brain structure and brain activity, and increased computational resources for detailed simulations with complex output. We report here (1) a novel approach to data sharing between collaborating scientists that brings together file system tools and cloud technologies, (2) a service implementing this approach, called NeuronDepot, and (3) an example application of the service to a complex use case in the neurosciences. The main drivers for our approach are to facilitate collaborations with a transparent, automated data flow that shields scientists from having to learn new tools or data structuring paradigms. Using NeuronDepot is simple: one-time data assignment from the originator and cloud based syncing-thus making experimental and modeling data available across the collaboration with minimum overhead. Since data sharing is cloud based, our approach opens up the possibility of using new software developments and hardware scalabitliy which are associated with elastic cloud computing. We provide an implementation that relies on existing synchronization services and is usable from all devices via a reactive web interface. We are motivating our solution by solving the practical problems of the GinJang project, a collaboration of three universities across eight time zones with a complex workflow encompassing data from electrophysiological recordings, imaging, morphological reconstructions, and simulations." ++++ diff --git a/content/publications/next-generation-cloud-databases-balancing-performance-sustainability-and-resource-management.md b/content/publications/next-generation-cloud-databases-balancing-performance-sustainability-and-resource-management.md new file mode 100644 index 0000000..5f89a20 --- /dev/null +++ b/content/publications/next-generation-cloud-databases-balancing-performance-sustainability-and-resource-management.md @@ -0,0 +1,10 @@ ++++ +title = "Next-Generation Cloud Databases: Balancing Performance, Sustainability, and Resource Management" +year = 2024 +authors = ["Michail Bachras", "Sofia Tijanic", "Shiquan Zhang", "Yuqiu Zhang", "Bryan Yan", "Lee Wing Piu", "Hans-Arno Jacobsen"] +venue = "2024 34th International Conference on Collaborative Advances in Software and COmputiNg (CASCON)" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/cascon62161.2024.10837926" +abstract = "This study enhances cloud database technologies for cyber-physical systems and IoT applications, addressing four key challenges: memory constraints, environmental impact, query performance, and serverless resource management. We introduce REMON, a network-based solution extending server memory capacity, and Krysha, a framework optimizing serverless microservices. Our research also includes an environmental impact assessment of analytical databases and explores hardware-conscious query tuning. Through extensive experiments, we demonstrate significant improvements in efficiency, scalability, and sustainability of cloud database systems. These advancements offer valuable solutions for managing large-scale data in complex digital ecosystems, with implications for both performance optimization and environmental considerations in cloud computing." ++++ diff --git a/content/publications/ninos-take-five-the-management-infrastructure-for-distributed-event-driven-workflows.md b/content/publications/ninos-take-five-the-management-infrastructure-for-distributed-event-driven-workflows.md new file mode 100644 index 0000000..1de8125 --- /dev/null +++ b/content/publications/ninos-take-five-the-management-infrastructure-for-distributed-event-driven-workflows.md @@ -0,0 +1,10 @@ ++++ +title = "NIÑOS take five: the management infrastructure for distributed event-driven workflows" +year = 2011 +authors = ["Siddarth Ganesan", "Young Yoon", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 5th ACM international conference on Distributed event-based system" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/2002259.2002286" +abstract = "Many workflows are inherently distributed over large-scale enterprise networks. These workflows involve many collaborating partners that interact in an autonomous and event-driven manner. Managing these workflows in an efficient and reliable manner is a challenging task. In this paper, we propose and evaluate the design for a distributed workflow management system that runs according to a set of protocols which utilize a content-based publish/subscribe messaging substrate. The novel features of our framework include elastic management of resources and flexible re-location of management entities to ensure cost-effective and low-latency management operations. Our experimental evaluation shows that the distributed and event-driven approach scales and maintains constant response time for varying management workloads. The average response time for management operations was reduced by 25% using our elastic management cluster approach compared to a fixed management cluster. Management requests were processed up to 10 times faster." ++++ diff --git a/content/publications/nofare-a-non-intrusive-facility-resource-monitoring-system.md b/content/publications/nofare-a-non-intrusive-facility-resource-monitoring-system.md new file mode 100644 index 0000000..8a95625 --- /dev/null +++ b/content/publications/nofare-a-non-intrusive-facility-resource-monitoring-system.md @@ -0,0 +1,9 @@ ++++ +title = "NoFaRe: A Non-Intrusive Facility Resource Monitoring System" +year = 2015 +authors = ["Matthias Kahl", "Christoph Goebel", "Anwar Ul Haq", "Thomas Kriechbaumer", "Hans-Arno Jacobsen"] +venue = "Lecture Notes in Computer Science; Energy Informatics" +publication_type = "Conference Paper" +research = [] +external_url = "https://doi.org/10.1007/978-3-319-25876-8_6" ++++ diff --git a/content/publications/omen-overlay-mending-for-topic-based-publish-subscribe-systems-under-churn.md b/content/publications/omen-overlay-mending-for-topic-based-publish-subscribe-systems-under-churn.md new file mode 100644 index 0000000..64e7ce4 --- /dev/null +++ b/content/publications/omen-overlay-mending-for-topic-based-publish-subscribe-systems-under-churn.md @@ -0,0 +1,10 @@ ++++ +title = "OMen: overlay mending for topic-based publish/subscribe systems under churn" +year = 2016 +authors = ["Chen Chen", "Roman Vitenberg", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 10th ACM International Conference on Distributed and Event-based Systems" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/2933267.2933305" +abstract = "We propose, OMen, a distributed system for dynamically maintaining overlays for topic-based publish/subscribe (pub/sub) systems. In particular, OMen supports churn-resistant construction of topic-connected overlays (TCO), which organizes all nodes interested in the same topic in a directly connected dissemination sub-overlay. While aiming at pub/sub deployments in data centers, OMen internally leverages selected peer-to-peer technologies, such as T-Man as the underlying topology maintenance protocol." ++++ diff --git a/content/publications/on-delivery-guarantees-in-distributed-content-based-publish-subscribe-systems.md b/content/publications/on-delivery-guarantees-in-distributed-content-based-publish-subscribe-systems.md new file mode 100644 index 0000000..06716ea --- /dev/null +++ b/content/publications/on-delivery-guarantees-in-distributed-content-based-publish-subscribe-systems.md @@ -0,0 +1,10 @@ ++++ +title = "On Delivery Guarantees in Distributed Content-Based Publish/Subscribe Systems" +year = 2020 +authors = ["Pooya Salehi", "Kaiwen Zhang", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 21st International Middleware Conference" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/3423211.3426400" +abstract = "Distributed overlay-based publish/subscribe systems provide a selective and scalable communication paradigm for connecting components of a distributed application. Existing overlay-based systems only guarantee delivery of notifications to clients that are already known by all brokers in the overlay. Nonetheless, due to the propagation delay, it takes time for a client's interests to be received by all brokers comprising the overlay. The message propagation delay and unclear delivery guarantees during this time increase the complexity of developing distributed applications based on the pub/sub paradigm. In this paper, we propose a collection of message processing and delivery guarantees that allows clients to clearly define the set of publications they receive. Based on our evaluation, these delivery guarantees can reduce buffering requirements on clients by up to 10 times, prevent missing notifications due to the propagation delay, and provide clients with primitive building blocks that simplify application development. We evaluate our proposed routing algorithms and show that a pub/sub system can provide the proposed delivery guarantees without increasing its resource requirements or hindering its throughput." ++++ diff --git a/content/publications/on-the-effects-of-distributed-electric-vehicle-network-utility-maximization-in-low-voltage-feeders.md b/content/publications/on-the-effects-of-distributed-electric-vehicle-network-utility-maximization-in-low-voltage-feeders.md new file mode 100644 index 0000000..42c59f7 --- /dev/null +++ b/content/publications/on-the-effects-of-distributed-electric-vehicle-network-utility-maximization-in-low-voltage-feeders.md @@ -0,0 +1,10 @@ ++++ +title = "On the Effects of Distributed Electric Vehicle Network Utility Maximization in Low Voltage Feeders" +year = 2017 +authors = ["Jose Rivera", "Hans-Arno Jacobsen"] +venue = "arXiv" +publication_type = "ArXiv Preprint" +research = [] +external_url = "https://arxiv.org/abs/1706.10074" +abstract = "The fast charging of Electric Vehicles (EVs) in distribution networks requires real-time EV charging control to avoid the overloading of grid components. Recent studies have proposed congestion control protocols, which result from distributed optimization solutions of the Network Utility Maximization (NUM) problem. While the NUM formulation allows the definition of distributed computations with closed form solutions, its simple model does not account for many of the feeders operational constraints. This puts the resulting control algorithms effectiveness into question. In this paper, we investigate the impact of implementing such algorithms for congestion control in low voltage feeders. We review the latest NUM based algorithms for real-time EV charging control, and evaluate their behavior and impact on the comprehensive IEEE European Low Voltage Test Feeder. Our results show that the EV NUM problem can effectively capture the relevant operational constraints, as long as ampacity violations are the main bottleneck. Moreover, the results demonstrate an advantage of the primal NUM solution over the more conventional dual NUM solution in preventing a system overload." ++++ diff --git a/content/publications/on-the-effects-of-signal-design-in-electric-vehicle-charging-using-vehicle-originating-signals.md b/content/publications/on-the-effects-of-signal-design-in-electric-vehicle-charging-using-vehicle-originating-signals.md new file mode 100644 index 0000000..0751372 --- /dev/null +++ b/content/publications/on-the-effects-of-signal-design-in-electric-vehicle-charging-using-vehicle-originating-signals.md @@ -0,0 +1,9 @@ ++++ +title = "On the effects of signal design in electric vehicle charging using vehicle-originating-signals" +year = 2016 +authors = ["Victor del Razo", "Christoph Goebel", "Hans-Arno Jacobsen"] +venue = "Computer Science - Research and Development" +publication_type = "Journal Article" +research = [] +external_url = "https://doi.org/10.1007/s00450-014-0286-4" ++++ diff --git a/content/publications/opengridmap-an-open-platform-for-inferring-power-grids-with-crowdsourced-data.md b/content/publications/opengridmap-an-open-platform-for-inferring-power-grids-with-crowdsourced-data.md new file mode 100644 index 0000000..9949d31 --- /dev/null +++ b/content/publications/opengridmap-an-open-platform-for-inferring-power-grids-with-crowdsourced-data.md @@ -0,0 +1,9 @@ ++++ +title = "OpenGridMap: An Open Platform for Inferring Power Grids with Crowdsourced Data" +year = 2015 +authors = ["José Rivera", "Christoph Goebel", "David Sardari", "Hans-Arno Jacobsen"] +venue = "Lecture Notes in Computer Science; Energy Informatics" +publication_type = "Conference Paper" +research = [] +external_url = "https://doi.org/10.1007/978-3-319-25876-8_15" ++++ diff --git a/content/publications/opengridmap-towards-automatic-power-grid-simulation-model-generation-from-crowdsourced-data.md b/content/publications/opengridmap-towards-automatic-power-grid-simulation-model-generation-from-crowdsourced-data.md new file mode 100644 index 0000000..f6b9755 --- /dev/null +++ b/content/publications/opengridmap-towards-automatic-power-grid-simulation-model-generation-from-crowdsourced-data.md @@ -0,0 +1,9 @@ ++++ +title = "OpenGridMap: towards automatic power grid simulation model generation from crowdsourced data" +year = 2017 +authors = ["José Rivera", "Johannes Leimhofer", "Hans-Arno Jacobsen"] +venue = "Computer Science - Research and Development" +publication_type = "Journal Article" +research = [] +external_url = "https://doi.org/10.1007/s00450-016-0317-4" ++++ diff --git a/content/publications/opportunistic-multipath-forwarding-in-content-based-publish-subscribe-overlays.md b/content/publications/opportunistic-multipath-forwarding-in-content-based-publish-subscribe-overlays.md new file mode 100644 index 0000000..2feec08 --- /dev/null +++ b/content/publications/opportunistic-multipath-forwarding-in-content-based-publish-subscribe-overlays.md @@ -0,0 +1,11 @@ ++++ +title = "Opportunistic Multipath Forwarding in Content-Based Publish/Subscribe Overlays" +year = 2012 +authors = ["Reza Sherafat Kazemzadeh", "Hans-Arno Jacobsen"] +venue = "Lecture Notes in Computer Science; Middleware 2012" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1007/978-3-642-35170-9_13" +abstract = "Fine-grained filtering capabilities prevalent in content-based Publish/Subscribe (pub/sub) overlays lead to scenarios in which publications pass through brokers with no matching local subscribers. Processing of publications at these pure forwarding brokers amounts to inefficient use of resources and should ideally be avoided. This paper develops an approach that largely mitigates this problem by building and adaptively maintaining a highly connected overlay mesh superimposed atop a low connectivity primary overlay network. While the primary network provides basic end-to-end forwarding routes, the mesh structure provides a rich set of alternative forwarding choices which can be used to bypass pure forwarding brokers. This provides unique opportunities for load balancing and congestion avoidance. Through extensive experimental evaluation on the SciNet cluster and PlanetLab, we compare the performance of our approach with that of conventional pub/sub algorithms as baseline. Our results indicate that our approach improves publication delivery delay and lowers network traffic while incurring negligible computational and bandwidth overhead. Furthermore, compared to the baseline, we observed significant gains of up to 115% in terms of system throughput." +abstract_license_url = "https://creativecommons.org/licenses/by/4.0/" ++++ diff --git a/content/publications/optimized-cluster-based-filtering-algorithm-for-graph-metadata.md b/content/publications/optimized-cluster-based-filtering-algorithm-for-graph-metadata.md new file mode 100644 index 0000000..f74f402 --- /dev/null +++ b/content/publications/optimized-cluster-based-filtering-algorithm-for-graph-metadata.md @@ -0,0 +1,9 @@ ++++ +title = "Optimized cluster-based filtering algorithm for graph metadata" +year = 2011 +authors = ["Haifeng Liu", "Zhaohui Wu", "Milenko Petrovic", "Hans-Arno Jacobsen"] +venue = "Information Sciences" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.1016/j.ins.2011.08.002" ++++ diff --git a/content/publications/optimizing-key-value-stores-for-hybrid-storage-architectures.md b/content/publications/optimizing-key-value-stores-for-hybrid-storage-architectures.md new file mode 100644 index 0000000..84e38b4 --- /dev/null +++ b/content/publications/optimizing-key-value-stores-for-hybrid-storage-architectures.md @@ -0,0 +1,10 @@ ++++ +title = "Optimizing key-value stores for hybrid storage architectures" +year = 2014 +authors = ["Prashanth Menon", "Tilmann Rabl", "Mohammad Sadoghi", "Hans-Arno Jacobsen"] +venue = "CASCON" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://dl.acm.org/citation.cfm?id=2735582" +abstract = "Flash-based solid state drives (SSDs) are increas-ingly becoming a popular choice as a storage de-vice within database management systems and key-value stores alike. SSDs offer fast throughput and low latency access to data, but their price-per-byte cost often makes them uneconomical for exclusive use, especially in the era of big data workloads. A common solution to this problem is to augment existing database systems by adding smaller SSDs that target only performance-critical areas. We be-lieve this hybrid approach to be a stop-gap solution. Rather than simply extending existing systems with SSDs, in this work we completely re-architect how a key-value database operates in a hybrid stor-age setting with both small but fast SSDs and slower but high-capacity HDDs. We formulate an accurate I/O cost model to study how popular key-value stores behave under several varying represen-tative workloads. Based on these studies and tak-ing a holistic approach, we design a system that dynamically optimizes the data layout and access strategy that leverages the strengths of each avail-able storage medium." ++++ diff --git a/content/publications/orca-observability-grounded-program-repair-for-microservice-incidents.md b/content/publications/orca-observability-grounded-program-repair-for-microservice-incidents.md new file mode 100644 index 0000000..6ad8762 --- /dev/null +++ b/content/publications/orca-observability-grounded-program-repair-for-microservice-incidents.md @@ -0,0 +1,10 @@ ++++ +title = "ORCA: Observability-Grounded Program Repair for Microservice Incidents" +year = 2026 +authors = ["Yuanchen Gao", "Yifang Tian", "Yiran Li", "Charles Zhang", "Hans-Arno Jacobsen"] +venue = "arXiv" +publication_type = "ArXiv Preprint" +research = ["data-management"] +external_url = "https://arxiv.org/abs/2608.17018" +abstract = "Microservice failures are often diagnosed from operational telemetry. However, automated program repair systems usually start from issue reports, localized code context, or failing tests. This mismatch leaves a gap between telemetry-based diagnosis and patch generation. We present ORCA, an observability-grounded APR pipeline for microservice incidents. ORCA first distills the differences in paired failure and reference telemetry into a fault signature, then uses the signature to identify candidate code and deployment-configuration locations. Repair graph agents and an Exploration agent generate unified-diff patch candidates from these locations. ORCA evaluates generated patches with a Telemetry-Grounded Patch Verifier that separates patch validity, syntactic and semantic correctness, test-oracle integrity, and telemetry replay. On a 575-case benchmark, ORCA outperforms all evaluated baselines in terms of cost-effectiveness. Results show that operational telemetry can be transformed from diagnostic evidence into actionable repair context: paired telemetry supports repair-oriented localization, while repair graph agents convert localized code and configuration evidence into constrained patch-generation context for the LLM. Telemetry-grounded verification then exposes repair outcomes that issue- or test-only evaluation would miss." ++++ diff --git a/content/publications/orchestrating-soa-using-requirement-specifications-and-domain-ontologies.md b/content/publications/orchestrating-soa-using-requirement-specifications-and-domain-ontologies.md new file mode 100644 index 0000000..51a1f1f --- /dev/null +++ b/content/publications/orchestrating-soa-using-requirement-specifications-and-domain-ontologies.md @@ -0,0 +1,9 @@ ++++ +title = "Orchestrating SOA Using Requirement Specifications and Domain Ontologies" +year = 2014 +authors = ["Manoj Bhat", "Chunyang Ye", "Hans-Arno Jacobsen"] +venue = "Lecture Notes in Computer Science; Service-Oriented Computing" +publication_type = "Conference Paper" +research = [] +external_url = "https://doi.org/10.1007/978-3-662-45391-9_30" ++++ diff --git a/content/publications/orderlesschain-a-crdt-based-bft-coordination-free-blockchain-without-global-order-of-transactions.md b/content/publications/orderlesschain-a-crdt-based-bft-coordination-free-blockchain-without-global-order-of-transactions.md new file mode 100644 index 0000000..12eba89 --- /dev/null +++ b/content/publications/orderlesschain-a-crdt-based-bft-coordination-free-blockchain-without-global-order-of-transactions.md @@ -0,0 +1,10 @@ ++++ +title = "OrderlessChain: A CRDT-based BFT Coordination-free Blockchain Without Global Order of Transactions" +year = 2023 +authors = ["Pezhman Nasirifard", "Ruben Mayer", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 24th International Middleware Conference on ZZZ" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/3590140.3629111" +abstract = "Existing permissioned blockchains often rely on coordination-based consensus protocols to ensure the safe execution of applications in a Byzantine environment. Furthermore, the protocols serialize the transactions by ordering them in a global order. The serializ-ability preserves the correctness of the application's state stored on the blockchain. However, coordination-based protocols limit the throughput and scalability and induce high latency. In contrast, application-level correctness requirements exist that are not dependent on the order of transactions, known as invariant-confluence (I-confluence). The I-confluent applications can execute transactions in a coordination-free manner, benefiting from the improved scalability compared to the coordination-based approaches. The safety and liveness of I-confluent applications are studied in non-Byzantine environments, but the correct execution of such applications remains a challenge in Byzantine coordination-free environments. We introduce OrderlessChain, a novel permissioned blockchain based on a novel BFT coordination-free protocol for the safe and live execution of I-confluent applications in a Byzantine environment. We implemented a prototype of our system, and our evaluation results show that our coordination-free approach performs significantly better than coordination-based blockchains." ++++ diff --git a/content/publications/out-of-core-edge-partitioning-at-linear-run-time.md b/content/publications/out-of-core-edge-partitioning-at-linear-run-time.md new file mode 100644 index 0000000..4c9ee70 --- /dev/null +++ b/content/publications/out-of-core-edge-partitioning-at-linear-run-time.md @@ -0,0 +1,10 @@ ++++ +title = "Out-of-Core Edge Partitioning at Linear Run-Time" +year = 2022 +authors = ["Ruben Mayer", "Kamil Orujzade", "Hans-Arno Jacobsen"] +venue = "2022 IEEE 38th International Conference on Data Engineering (ICDE)" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icde53745.2022.00242" +abstract = "Graph edge partitioning is an important prepro-cessing step to optimize distributed computing jobs on graph-structured data. The edge set of a given graph is split into $k$ equally-sized partitions, such that the replication of vertices across partitions is minimized. Out-of-core edge partitioning algorithms are able to tackle the problem with low memory over-head. Existing out-of-core algorithms mainly work in a streaming manner and can be grouped into two types. While stateless streaming edge partitioning is fast and yields low partitioning quality, stateful streaming edge partitioning yields better quality, but is expensive, as it requires a scoring function to be evaluated for every edge on every partition, leading to a time complexity of O(|E| \\*k). In this paper, we propose 2PS-L, a novel out-of-core edge partitioning algorithm that builds upon the stateful streaming model, but achieves linear run-time i.e.,O(|E|)). 2PS-L consists of two phases. In the first phase, vertices are separated into clusters by a lightweight streaming clustering algorithm. In the second phase, the graph is re-streamed and vertex clustering from the first phase is exploited to reduce the search space of graph partitioning to only two target partitions for every edge. Our evaluations show that 2PS-L can achieve better partitioning quality than existing stateful streaming edge partitioners while having a much lower run-time. As a consequence, the total run-time of partitioning and subsequent distributed graph processing can be significantly reduced." ++++ diff --git a/content/publications/overcoming-data-scarcity-through-transfer-learning-in-co2-based-building-occupancy-detection.md b/content/publications/overcoming-data-scarcity-through-transfer-learning-in-co2-based-building-occupancy-detection.md new file mode 100644 index 0000000..fdbe28e --- /dev/null +++ b/content/publications/overcoming-data-scarcity-through-transfer-learning-in-co2-based-building-occupancy-detection.md @@ -0,0 +1,10 @@ ++++ +title = "Overcoming Data Scarcity through Transfer Learning in CO2-Based Building Occupancy Detection" +year = 2023 +authors = ["Manuel Weber", "Farzan Banihashemi", "Peter Mandl", "Hans-Arno Jacobsen", "Ruben Mayer"] +venue = "Proceedings of the 10th ACM International Conference on Systems for Energy-Efficient Buildings, Cities, and Transportation" +publication_type = "Conference Paper" +research = ["distributed-machine-learning"] +external_url = "https://doi.org/10.1145/3600100.3623718" +abstract = "Knowing indoor occupancy states is crucial for energy optimization in buildings. While neural networks can effectively be used to detect occupancy based on carbon dioxide measurements, their application is impeded by the need for sufficient labeled training data. In this study, we analyze the prediction performance of three different transfer learning (TL) methods leveraging target room data jointly with data from other rooms. The methods include (1) pretraining and fine-tuning, (2) layer freezing, and (3) domain-adversarial learning. Using data from five real-world rooms and one simulated room, including multiple room types, we provide the most extensive evaluation of TL in the field of occupancy prediction from environmental variables to date. This work’s contribution further includes the architecture and hyperparameters of a deep CNN-LSTM model for CO2-based occupancy detection. Our results indicate that TL effectively reduces the required amount of target room data. Moreover, while previous literature was focused on pretraining with related real-world data, we show that similar performance can be achieved by the more practical approach of leveraging simulated data." ++++ diff --git a/content/publications/overlay-design-for-topic-based-publish-subscribe-under-node-degree-constraints.md b/content/publications/overlay-design-for-topic-based-publish-subscribe-under-node-degree-constraints.md new file mode 100644 index 0000000..e4daa25 --- /dev/null +++ b/content/publications/overlay-design-for-topic-based-publish-subscribe-under-node-degree-constraints.md @@ -0,0 +1,10 @@ ++++ +title = "Overlay Design for Topic-Based Publish/Subscribe under Node Degree Constraints" +year = 2016 +authors = ["Chen Chen", "Yoav Tock", "Hans-Arno Jacobsen"] +venue = "2016 IEEE 36th International Conference on Distributed Computing Systems (ICDCS)" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icdcs.2016.79" +abstract = "It is important to build overlays for topic-based publish/subscribe (pub/sub) under resource constraints. In a topic-connected overlay (TCO), each topic t induces a connected sub-overlay among all nodes interested in t. Existing work merely consider how to optimize a complete TCO and implicitly commit the unrealistic assumption of unlimited resources. In contrast, we make maximum use of restricted node degree budgets to build a partial TCO. We formalize the notion of TCO support to capture the quality of the pub/sub overlay. Furthermore, we demonstrate that partial TCOs usually exhibit significantly better cost-effectiveness in practice. We propose two problems of maximizing TCO support in a partial TCO: (1) PTCOA with a bounded average node degree and (2) PTCOM under the maximum node degree constraint. We design two greedy algorithms, which achieve the constant approximation ratios of (1-e-1) for PTCOA and (1-e-1/6) for PTCOM, respectively. Empirical evaluation demonstrates the scalability of our algorithms under a variety of pub/sub workloads. Given practical data sets extracted from Facebook and Twitter, our algorithms produce an 80% TCO with fewer than 20% of the node degree budget as a complete TCO. We also show experimentally that it is promising to design decentralized protocols to compute a partial TCO for pub/sub." ++++ diff --git a/content/publications/panjoin-a-partition-based-adaptive-stream-join.md b/content/publications/panjoin-a-partition-based-adaptive-stream-join.md new file mode 100644 index 0000000..d83ed0a --- /dev/null +++ b/content/publications/panjoin-a-partition-based-adaptive-stream-join.md @@ -0,0 +1,10 @@ ++++ +title = "PanJoin: A Partition-based Adaptive Stream Join" +year = 2018 +authors = ["Fei Pan", "Hans-Arno Jacobsen"] +venue = "arXiv" +publication_type = "ArXiv Preprint" +research = ["data-management"] +external_url = "https://arxiv.org/abs/1811.05065" +abstract = "In stream processing, stream join is one of the critical sources of performance bottlenecks. The sliding-window-based stream join provides a precise result but consumes considerable computational resources. The current solutions lack support for the join predicates on large windows. These algorithms and their hardware accelerators are either limited to equi-join or use a nested loop join to process all the requests. In this paper, we present a new algorithm called PanJoin which has high throughput on large windows and supports both equi-join and non-equi-join. PanJoin implements three new data structures to reduce computations during the probing phase of stream join. We also implement the most hardware-friendly data structure, called BI-Sort, on FPGA. Our evaluation shows that PanJoin outperforms several recently proposed stream join methods by more than 1000x, and it also adapts well to highly skewed data." ++++ diff --git a/content/publications/parallel-index-based-stream-join-on-a-multicore-cpu.md b/content/publications/parallel-index-based-stream-join-on-a-multicore-cpu.md new file mode 100644 index 0000000..ddd25e3 --- /dev/null +++ b/content/publications/parallel-index-based-stream-join-on-a-multicore-cpu.md @@ -0,0 +1,10 @@ ++++ +title = "Parallel Index-based Stream Join on a Multicore CPU" +year = 2020 +authors = ["Amirhesam Shahvarani", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 2020 ACM SIGMOD International Conference on Management of Data" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/3318464.3380576" +abstract = "Indexing sliding window content to enhance the performance of streaming queries can be greatly improved by utilizing the computational capabilities of a multicore processor. Conventional indexing data structures optimized for frequent search queries on a prestored dataset do not meet the demands of indexing highly dynamic data as in streaming environments. In this paper, we introduce an index data structure, called the partitioned in-memory merge tree, to address the challenges that arise when indexing highly dynamic data, which are common in streaming settings. Utilizing the specific pattern of streaming data and the distribution of queries, we propose a low-cost and effective concurrency control mechanism to meet the demands of high-rate update queries. To complement the index, we design an algorithm to realize a parallel index-based stream join that exploits the computational power of multicore processors. Our experiments using an octa-core processor show that our parallel stream join achieves up to 5.5 times higher throughput than a single-threaded approach." ++++ diff --git a/content/publications/parametrized-quantum-circuit-learning-for-quantum-chemical-applications.md b/content/publications/parametrized-quantum-circuit-learning-for-quantum-chemical-applications.md new file mode 100644 index 0000000..75fddc8 --- /dev/null +++ b/content/publications/parametrized-quantum-circuit-learning-for-quantum-chemical-applications.md @@ -0,0 +1,10 @@ ++++ +title = "Parametrized Quantum Circuit Learning for Quantum Chemical Applications" +year = 2026 +authors = ["Grier M. Jones", "Viki Kumar Prasad", "Ulrich Fekl", "Hans-Arno Jacobsen"] +venue = "Journal of Chemical Information and Modeling" +publication_type = "Journal Article" +research = ["quantum-computing-systems", "distributed-machine-learning"] +external_url = "https://doi.org/10.1021/acs.jcim.6c00376" +abstract = "In the field of quantum machine learning (QML), parametrized quantum circuits (PQCs)─constructed using a combination of fixed and tunable quantum gates─provide a promising hybrid framework for tackling complex machine learning problems. Despite numerous proposed applications, there remains limited exploration of data sets relevant to quantum chemistry. In this study, we investigate the potential benefits and limitations of PQCs on two chemically meaningful data sets: (1) the BSE49 data set, containing bond separation energies for 49 different classes of chemical bonds, and (2) a data set of water conformers, where coupled-cluster singles and doubles (CCSD) wave functions are predicted from lower-level electronic structure methods using the data-driven coupled-cluster (DDCC) approach. We construct a comprehensive set of 168 PQCs by combining 14 data encoding strategies with 12 variational ansätze, and evaluate their performance on circuits with 5 and 16 qubits. Our initial analysis examines the impact of circuit structure on model performance using state-vector simulations. We then explore how circuit depth and training set size influence model performance. Finally, we assess the performance of the best-performing PQCs on current quantum hardware, using both noisy simulations (\"fake\" backends) and real quantum devices. Our findings underscore the challenges of applying PQCs to chemically relevant problems that are straightforward for classical machine learning methods but remain nontrivial for quantum approaches." ++++ diff --git a/content/publications/parparaw-massively-parallel-parsing-of-delimiter-separated-raw-data.md b/content/publications/parparaw-massively-parallel-parsing-of-delimiter-separated-raw-data.md new file mode 100644 index 0000000..b68e463 --- /dev/null +++ b/content/publications/parparaw-massively-parallel-parsing-of-delimiter-separated-raw-data.md @@ -0,0 +1,10 @@ ++++ +title = "ParPaRaw: Massively Parallel Parsing of Delimiter-Separated Raw Data" +year = 2020 +authors = ["Elias Stehle", "Hans-Arno Jacobsen"] +venue = "Proceedings of the VLDB Endowment" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.14778/3377369.3377372" +abstract = "Parsing is essential for a wide range of use cases, such as stream processing, bulk loading, and in-situ querying of raw data. Yet, the compute-intense step often constitutes a major bottleneck in the data ingestion pipeline, since parsing of inputs that require more involved parsing rules is challenging to parallelise. This work proposes a massively parallel algorithm for parsing delimiter-separated data formats on GPUs. Other than the state-of-the-art, the proposed approach does not require an initial sequential pass over the input to determine a thread's parsing context. That is, how a thread, beginning somewhere in the middle of the input, should interpret a certain symbol (e.g., whether to interpret a comma as a delimiter or as part of a larger string enclosed in double-quotes). Instead of tailoring the approach to a single format, we are able to perform a massively parallel finite state machine (FSM) simulation, which is more flexible and powerful, supporting more expressive parsing rules with general applicability. Achieving a parsing rate of as much as 14.2 GB/s, our experimental evaluation on a GPU with 3 584 cores shows that the presented approach is able to scale to thousands of cores and beyond. With an end-to-end streaming approach, we are able to exploit the full-duplex capabilities of the PCIe bus and hide latency from data transfers. Considering the end-to-end performance, the algorithm parses 4.8 GB in as little as 0.44 seconds, including data transfers." ++++ diff --git a/content/publications/partition-tolerant-distributed-publish-subscribe-systems.md b/content/publications/partition-tolerant-distributed-publish-subscribe-systems.md new file mode 100644 index 0000000..8d53c36 --- /dev/null +++ b/content/publications/partition-tolerant-distributed-publish-subscribe-systems.md @@ -0,0 +1,10 @@ ++++ +title = "Partition-Tolerant Distributed Publish/Subscribe Systems" +year = 2011 +authors = ["Reza Sherafat Kazemzadeh", "Hans-Arno Jacobsen"] +venue = "2011 IEEE 30th International Symposium on Reliable Distributed Systems" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/srds.2011.21" +abstract = "In this paper, we develop reliable distributed publish/subscribe algorithms that can tolerate concurrent failure of up to d broker machines or communication links. In our approach, d is a configuration parameter which determines the level of fault-tolerance of the system and reliability refers to exactly-once and per-source, in-order delivery of publications to clients with matching subscriptions. We propose protocols to address three problems in presence of broker or link failures: (i) subscription propagation, (ii) publication forwarding, and (iii) broker recovery. Finally, we study the effectiveness of our approach when the number of concurrent failures exceeds d. Through large-scale experimental evaluations with up to 500 brokers, we demonstrate that a system configured with a modest value of d = 3 is able to reliably deliver 97% of publications in presence of failure of up to 17% of its brokers." ++++ diff --git a/content/publications/partitioner-selection-with-ease.md b/content/publications/partitioner-selection-with-ease.md index a219655..26ed347 100644 --- a/content/publications/partitioner-selection-with-ease.md +++ b/content/publications/partitioner-selection-with-ease.md @@ -6,6 +6,6 @@ venue = "2023 IEEE 39th International Conference on Data Engineering (ICDE)" publication_type = "Conference Paper" research = ["distributed-machine-learning", "data-management"] tags = ["graph-systems", "graph-processing", "auto-tuning"] -summary = "Conference paper on automatic graph-partitioner selection for distributed graph processing." -external_url = "https://ieeexplore.ieee.org/abstract/document/10184652/" +external_url = "https://doi.org/10.1109/icde55515.2023.00185" +abstract = "For distributed graph processing on massive graphs, a graph is partitioned into multiple equally-sized parts which are distributed among machines in a compute cluster. In the last decade, many partitioning algorithms have been developed which differ from each other with respect to the partitioning quality, the run-time of the partitioning and the type of graph for which they work best. The plethora of graph partitioning algorithms makes it a challenging task to select a partitioner for a given scenario. Different studies exist that provide qualitative insights into the characteristics of graph partitioning algorithms that support a selection. However, in order to enable automatic selection, a quantitative prediction of the partitioning quality, the partitioning run-time and the run-time of subsequent graph processing jobs is needed. In this paper, we propose a machine learning-based approach to provide such a quantitative prediction for different types of edge partitioning algorithms and graph processing workloads. We show that training based on generated graphs achieves high accuracy, which can be further improved when using real-world data. Based on the predictions, the automatic selection reduces the end-to-end run-time on average by 11.1% compared to a random selection, by 17.4% compared to selecting the partitioner that yields the lowest cut size, and by 29.1% compared to the worst strategy, respectively. Furthermore, in 35.7% of the cases, the best strategy was selected." +++ diff --git a/content/publications/performance-evaluation-and-optimization-of-multi-dimensional-indexes-in-hive.md b/content/publications/performance-evaluation-and-optimization-of-multi-dimensional-indexes-in-hive.md new file mode 100644 index 0000000..288296f --- /dev/null +++ b/content/publications/performance-evaluation-and-optimization-of-multi-dimensional-indexes-in-hive.md @@ -0,0 +1,10 @@ ++++ +title = "Performance Evaluation and Optimization of Multi-Dimensional Indexes in Hive" +year = 2018 +authors = ["Yue Liu", "Shuai Guo", "Songlin Hu", "Tilmann Rabl", "Hans-Arno Jacobsen", "Jintao Li", "Jiye Wang"] +venue = "IEEE Transactions on Services Computing" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.1109/tsc.2016.2594778" +abstract = "Apache Hive has been widely used for big data processing over large scale clusters by many companies. It provides a declarative query language called HiveQL. The efficiency of filtering out query-irrelevant data from HDFS closely affects the performance of query processing. This is especially true for multi-dimensional, high-selective, and few columns involving queries, which provides sufficient information to reduce the amount of bytes read. Indexing (Compact Index, Aggregate Index, Bitmap Index, DGFIndex, and the index in ORC file) and columnar storage (RCFile, ORC file, and Parquet) are powerful techniques to achieve this. However, it is not trivial to choosing a suitable index and columnar storage based on data and query features. In this paper, we compare the data filtering performance of the above indexes with different columnar storage formats by conducting comprehensive experiments using uniform and skew TPC-H data sets and various multi-dimensional queries, and suggest the best practices of improving multi-dimensional queries in Hive under different conditions." ++++ diff --git a/content/publications/planning-the-transformation-of-overlays.md b/content/publications/planning-the-transformation-of-overlays.md new file mode 100644 index 0000000..527e2fc --- /dev/null +++ b/content/publications/planning-the-transformation-of-overlays.md @@ -0,0 +1,10 @@ ++++ +title = "Planning the transformation of overlays" +year = 2016 +authors = ["Young Yoon", "Nathan Robinson", "Vinod Muthusamy", "Sheila A. McIlraith", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 31st Annual ACM Symposium on Applied Computing" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/2851613.2851639" +abstract = "Reconfiguring a topology is an important technique to sustain high efficiency and robustness of an overlay. However, the problem of transforming the overlay from an old topology to a newly refined one, at runtime, has received relatively little attention. The key challenge is to minimize the disruption that can be caused by topology transformation operations. Excessive disruption can be costly thus hamper the decision to migrate to a better topology. To address this issue, we solve a problem of finding an appropriate sequence of steps to transform a topology that incurs the least service disruption. We call this the incremental topology transformation (ITT) problem. ITT can be formulated well as an automated planning problem and can be solved with numerous off-the-shelf planning algorithms. However, we found that state-of-the-art domain-independent planning techniques can not scale to solve large ITT problem instances. This shortcoming motivated us to develop a suite of planners that use novel domain-specific heuristics to guide the search for a solution. Our empirical evaluation shows that our planners offer a viable solution to a diversity of ITT problems." ++++ diff --git a/content/publications/popsub-improving-resource-utilization-in-distributed-content-based-publish-subscribe-systems.md b/content/publications/popsub-improving-resource-utilization-in-distributed-content-based-publish-subscribe-systems.md new file mode 100644 index 0000000..e0de693 --- /dev/null +++ b/content/publications/popsub-improving-resource-utilization-in-distributed-content-based-publish-subscribe-systems.md @@ -0,0 +1,10 @@ ++++ +title = "PopSub: Improving Resource Utilization in Distributed Content-based Publish/Subscribe Systems" +year = 2017 +authors = ["Pooya Salehi", "Kaiwen Zhang", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 11th ACM International Conference on Distributed and Event-based Systems" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/3093742.3093915" +abstract = "Distributed content-based publish/subscribe systems provide a selective, scalable, and decentralized approach to data dissemination. In a pub/sub overlay network, hop-by-hop routing allows brokers to correctly forward messages without requiring global knowledge. However, this model causes brokers to forward publications without knowing the volume and distance of matching subscribers, which can result in inefficient resource utilization. In order to raise the scalability of pub/sub, we introduce Popularity-Based Publication Routing for Content-based Pub/Sub (PopSub), which is specifically designed to raise the resource utilization efficiency. We define a utilization metric to measure the impact of forwarding a publication on the overall delivery of the system. Furthermore, we propose a new publication routing algorithm that takes into account broker resources and publication popularity among subscribers. Lastly, we propose three approaches to handle unpopular publications. Based on our evaluations, using real-world workloads and traces, PopSub is able to improve resource efficiency of the brokers by up to 62%, and reduce delivery latency by up to 57% under high load." ++++ diff --git a/content/publications/position-let-s-develop-data-probes-to-fundamentally-understand-how-data-affects-llm-performance.md b/content/publications/position-let-s-develop-data-probes-to-fundamentally-understand-how-data-affects-llm-performance.md new file mode 100644 index 0000000..d65cc36 --- /dev/null +++ b/content/publications/position-let-s-develop-data-probes-to-fundamentally-understand-how-data-affects-llm-performance.md @@ -0,0 +1,11 @@ ++++ +title = "Position: Let's Develop Data Probes to Fundamentally Understand How Data Affects LLM Performance" +year = 2026 +authors = ["Shiqiang Wang", "Herbert Woisetschläger", "Hans-Arno Jacobsen", "Mingyue Ji"] +venue = "International Conference on Machine Learning (ICML)" +publication_type = "Conference Paper" +research = ["distributed-machine-learning"] +external_url = "https://icml.cc/virtual/2026/poster/67154" +abstract = "Data is fundamental to large language models (LLMs). However, understanding of what makes certain data useful for different stages of an LLM workflow, including training, tuning, alignment, in-context learning, etc., and why, remains an open question. Current approaches rely heavily on extensive experimentation with large public datasets to obtain empirical heuristics for data filtering and dataset construction. These approaches are compute intensive and lack a principled way of understanding the essence of how specific data characteristics drive LLM behavior. In this position paper, we advocate for the need of developing systematic methodologies for generating synthetic sequences from appropriately defined random processes, with the goal that these sequences can reveal useful characteristics when they are used in one or multiple stages of the LLM workflow. We refer to such sequences as data probes. By observing LLM behavior on data probes, researchers can systematically conduct studies on how data characteristics influence model performance, generalization, and robustness. The probing sequences exhibit statistical properties that can be viewed using theoretical concepts, such as typical sets, which are generalized to describe the behaviors of LLMs. This data-probe approach provides a pathway for uncovering foundational insights into the role of data in LLM training and inference, beyond empirical heuristics." +abstract_license_url = "https://creativecommons.org/licenses/by/4.0/" ++++ diff --git a/content/publications/predict-predictive-dictionary-maintenance-for-message-compression-in-publish-subscribe.md b/content/publications/predict-predictive-dictionary-maintenance-for-message-compression-in-publish-subscribe.md new file mode 100644 index 0000000..26f5124 --- /dev/null +++ b/content/publications/predict-predictive-dictionary-maintenance-for-message-compression-in-publish-subscribe.md @@ -0,0 +1,10 @@ ++++ +title = "PreDict: Predictive Dictionary Maintenance for Message Compression in Publish/Subscribe" +year = 2018 +authors = ["Christoph Doblander", "Arash Khatayee", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 19th International Middleware Conference" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/3274808.3274822" +abstract = "Data usage is a significant concern, particularly in smartphone applications, M2M communications and for Internet of Things (IoT) applications. Messages in these domains are often exchanged with a backend infrastructure using publish/subscribe (pub/sub). Shared dictionary compression has been shown to reduce data usage in pub/sub networks beyond that obtained using well-known techniques, such as DEFLATE, gzip and delta encoding, but such compression requires manual configuration, which increases the operational complexity." ++++ diff --git a/content/publications/predictive-publish-subscribe-matching.md b/content/publications/predictive-publish-subscribe-matching.md new file mode 100644 index 0000000..946fd83 --- /dev/null +++ b/content/publications/predictive-publish-subscribe-matching.md @@ -0,0 +1,10 @@ ++++ +title = "Predictive publish/subscribe matching" +year = 2010 +authors = ["Vinod Muthusamy", "Haifeng Liu", "Hans-Arno Jacobsen"] +venue = "Proceedings of the Fourth ACM International Conference on Distributed Event-Based Systems" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/1827418.1827423" +abstract = "A new publish/subscribe capability is presented: the ability to predict the likelihood that a subscription will be matched at some point in the future. Knowing that some phenomenon of interest is about to take place, applications can take proactive steps to prevent the situation from occurring altogether, or speculatively begin reacting to the event even before it has transpired. A publish/subscribe matching algorithm is developed in which composite subscriptions consisting of temporal and logical operators are efficiently represented by a set of finite state machines and rules. The algorithm trains a Markov model to an application's event workload, and predicts the probability that a given subscription will match within a window in the future event stream. Evaluations demonstrate that the memory and processing costs of the algorithm scale well with the number of subscriptions, and the prediction precision is high, especially when the workload characteristics do not change rapidly. Furthermore, a comparison with a hand-crafted Markov model using real data traces shows that the algorithm consumes much less memory and processing power, yet still delivers prediction precision that approaches that of the hand-crafted model. This is especially impressive since the algorithms lack any of the domain expertise embedded in the hand-crafted model." ++++ diff --git a/content/publications/prestigebft-revolutionizing-view-changes-in-bft-consensus-algorithms-with-reputation-mechanisms.md b/content/publications/prestigebft-revolutionizing-view-changes-in-bft-consensus-algorithms-with-reputation-mechanisms.md new file mode 100644 index 0000000..47971c4 --- /dev/null +++ b/content/publications/prestigebft-revolutionizing-view-changes-in-bft-consensus-algorithms-with-reputation-mechanisms.md @@ -0,0 +1,10 @@ ++++ +title = "PrestigeBFT: Revolutionizing View Changes in BFT Consensus Algorithms with Reputation Mechanisms" +year = 2024 +authors = ["Gengrui Zhang", "Fei Pan", "Sofia Tijanic", "Hans-Arno Jacobsen"] +venue = "2024 IEEE 40th International Conference on Data Engineering (ICDE)" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icde60146.2024.00156" +abstract = "Passive view-change protocols are widely employed in BFT algorithms; however, they present the risks of selecting unavailable or slow servers as leaders. To tackle these challenges, we propose PrestigeBFT, a novel BFT consensus algorithm that incorporates an active view-change protocol with reputation mechanisms. PrestigeBFT evaluates a server's reputation based on its past behavior and elects more reputable servers as leaders. Our reputation mechanism incentivizes protocol-abiding behavior while penalizing faulty servers by imposing computational work. PrestigeBFT significantly enhances system availability and efficiency by avoiding unavailable or slow servers being assigned as leaders. Under normal operation, PrestigeBFT achieves $5\\times$ higher throughput than the baseline that uses passive view-change protocols. In addition, PrestigeBFT's throughput remains unaffected under benign faults and witnesses only a 24% drop under a variety of Byzantine faults, whereas the baseline throughput drops by 62% and 69%, respectively. In the long run, while the baseline's availability struggles at 37%, PrestigeBFT progressively improves its availability to over 90%." ++++ diff --git a/content/publications/prism-is-research-in-aspect-mining.md b/content/publications/prism-is-research-in-aspect-mining.md new file mode 100644 index 0000000..6993c2b --- /dev/null +++ b/content/publications/prism-is-research-in-aspect-mining.md @@ -0,0 +1,9 @@ ++++ +title = "PRISM is research in aSpect mining" +year = 2004 +authors = ["Charles Zhang", "Hans-Arno Jacobsen"] +venue = "Companion to the 19th annual ACM SIGPLAN conference on Object-oriented programming systems, languages, and applications" +publication_type = "Conference Paper" +research = [] +external_url = "https://doi.org/10.1145/1028664.1028676" ++++ diff --git a/content/publications/probabilistic-design-of-parametrized-quantum-circuits-through-local-gate-modifications.md b/content/publications/probabilistic-design-of-parametrized-quantum-circuits-through-local-gate-modifications.md new file mode 100644 index 0000000..173acf1 --- /dev/null +++ b/content/publications/probabilistic-design-of-parametrized-quantum-circuits-through-local-gate-modifications.md @@ -0,0 +1,10 @@ ++++ +title = "Probabilistic Design of Parametrized Quantum Circuits through Local Gate Modifications" +year = 2026 +authors = ["Grier M. Jones", "Aviraj Newatia", "Alexander Lao", "Aditya K. Rao", "Viki Kumar Prasad", "Hans-Arno Jacobsen"] +venue = "arXiv" +publication_type = "ArXiv Preprint" +research = ["quantum-computing-systems"] +external_url = "https://arxiv.org/abs/2602.12465" +abstract = "Within quantum machine learning, parametrized quantum circuits provide flexible quantum models, but their performance is often highly task-dependent, making manual circuit design challenging. Alternatively, quantum architecture search algorithms have been proposed to automate the discovery of task-specific parametrized quantum circuits using systematic frameworks. In this work, we propose an evolution-inspired heuristic quantum architecture search algorithm, which we refer to as the local quantum architecture search. The goal of the local quantum architecture search algorithm is to optimize parametrized quantum circuit architectures through a local, probabilistic search over a fixed set of gate-level actions applied to existing circuits. We evaluate the local quantum architecture search algorithm on two synthetic function-fitting regression tasks and two quantum chemistry regression datasets, including the BSE49 dataset of bond separation energies for first- and second-row elements and a dataset of water conformers generated using the data-driven coupled-cluster approach. Using state-vector simulation, our results highlight the applicability of local quantum architecture search algorithm for identifying competitive circuit architectures with desirable performance metrics. Lastly, we analyze the properties of the discovered circuits and demonstrate the deployment of the best-performing model on state-of-the-art quantum hardware." ++++ diff --git a/content/publications/process-discovery-from-dependence-complete-event-logs.md b/content/publications/process-discovery-from-dependence-complete-event-logs.md new file mode 100644 index 0000000..b0cc1db --- /dev/null +++ b/content/publications/process-discovery-from-dependence-complete-event-logs.md @@ -0,0 +1,10 @@ ++++ +title = "Process Discovery from Dependence-Complete Event Logs" +year = 2016 +authors = ["Wei Song", "Hans-Arno Jacobsen", "Chunyang Ye", "Xiaoxing Ma"] +venue = "IEEE Transactions on Services Computing" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.1109/tsc.2015.2426181" +abstract = "Process mining, especially process discovery, has been utilized to extract process models from event logs. One challenge faced by process discovery is to identify concurrency effectively. State-of-the-art approaches employ activity orders in traces to undertake process discovery and they require stringent completeness notions of event logs. Thus, they may fail to extract appropriate processes when event logs cannot meet the completeness criteria. To address this problem, we propose in this paper a novel technique which leverages activity dependences in traces. Based on the observation that activities with no dependencies can be executed in parallel, our technique is in a position to discover processes with concurrencies even if the logs fail to meet the completeness criteria. That is, our technique calls for a weaker notion of completeness. We evaluate our technique through experiments on both real-world and synthetic event logs, and the conformance checking results demonstrate the effectiveness of our technique and its relative advantages compared with state-of-the-art approaches." ++++ diff --git a/content/publications/processing-big-events-with-showers-and-streams.md b/content/publications/processing-big-events-with-showers-and-streams.md new file mode 100644 index 0000000..5507b7a --- /dev/null +++ b/content/publications/processing-big-events-with-showers-and-streams.md @@ -0,0 +1,9 @@ ++++ +title = "Processing Big Events with Showers and Streams" +year = 2014 +authors = ["Christoph Doblander", "Tilmann Rabl", "Hans-Arno Jacobsen"] +venue = "Lecture Notes in Computer Science; Specifying Big Data Benchmarks" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1007/978-3-642-53974-9_6" ++++ diff --git a/content/publications/processing-proximity-relations-in-road-networks.md b/content/publications/processing-proximity-relations-in-road-networks.md new file mode 100644 index 0000000..46ba546 --- /dev/null +++ b/content/publications/processing-proximity-relations-in-road-networks.md @@ -0,0 +1,10 @@ ++++ +title = "Processing proximity relations in road networks" +year = 2010 +authors = ["Zhengdao Xu", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 2010 ACM SIGMOD International Conference on Management of data" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/1807167.1807196" +abstract = "Applications ranging from location-based services to multi-player online gaming require continuous query support to monitor, track, and detect events of interest among sets of moving objects. Examples are alerting capabilities for detecting whether the distance, the travel cost, or the travel time among a set of moving objects exceeds a threshold. These types of queries are driven by continuous streams of location updates, simultaneously evaluated over many queries." ++++ diff --git a/content/publications/profitability-of-residential-battery-energy-storage-combined-with-solar-photovoltaics.md b/content/publications/profitability-of-residential-battery-energy-storage-combined-with-solar-photovoltaics.md new file mode 100644 index 0000000..f033a13 --- /dev/null +++ b/content/publications/profitability-of-residential-battery-energy-storage-combined-with-solar-photovoltaics.md @@ -0,0 +1,10 @@ ++++ +title = "Profitability of Residential Battery Energy Storage Combined with Solar Photovoltaics" +year = 2017 +authors = ["Christoph Goebel", "Vicky Cheng", "Hans-Arno Jacobsen"] +venue = "Energies" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.3390/en10070976" +abstract = "Lithium-ion (Li-Ion) batteries are increasingly being considered as bulk energy storage in grid applications. One such application is residential energy storage combined with solar photovoltaic (PV) panels to enable higher self-consumption rates, which has become financially more attractive recently due to decreasing feed-in subsidies. Although residential energy storage solutions are commercially mature, it remains unclear which system configurations and circumstances, including aggregator-based applications such as the provision of ancillary services, lead to profitable consumer investments. Therefore, we conduct an extensive simulation study that is able to jointly capture these aspects. Our results show that, at current battery module prices, even optimal system configurations still do not lead to profitable investments into Li-Ion batteries if they are merely used as a buffer for solar energy. The first settings in which they will become profitable, as prices are further declining, will be larger households at locations with higher average levels of solar irradiance. If the batteries can be remote-controlled by an aggregator to provide overnight negative reserve, their profitability increases significantly." ++++ diff --git a/content/publications/prosecutor-an-efficient-bft-consensus-algorithm-with-behavior-aware-penalization-against-byzantine-attacks.md b/content/publications/prosecutor-an-efficient-bft-consensus-algorithm-with-behavior-aware-penalization-against-byzantine-attacks.md new file mode 100644 index 0000000..ca499f0 --- /dev/null +++ b/content/publications/prosecutor-an-efficient-bft-consensus-algorithm-with-behavior-aware-penalization-against-byzantine-attacks.md @@ -0,0 +1,10 @@ ++++ +title = "Prosecutor: an efficient BFT consensus algorithm with behavior-aware penalization against Byzantine attacks" +year = 2021 +authors = ["Gengrui Zhang", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 22nd International Middleware Conference" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/3464298.3484503" +abstract = "Current leader-based Byzantine fault-tolerant (BFT) protocols aim to improve the efficiency for achieving consensus while tolerating failures; however, Byzantine servers are able to repeatedly impair BFT systems as faulty servers launch attacks without costs. In this paper, leveraging Proof-of-Work and Raft, we propose a new BFT consensus protocol called Prosecutor that dynamically penalizes suspected faulty behavior and suppresses Byzantine servers over time. Prosecutor obstructs Byzantine servers from being elected in leader election by imposing hash computation on new election campaigns. Furthermore, Prosecutor applies message authentication to achieve secure log replication and maintains a similar message-passing scheme as Raft. The evaluation results show that the penalization mechanism progressively suppresses and marginalizes Byzantine servers if they repeatedly launch malicious attacks." ++++ diff --git a/content/publications/prospects-of-appliance-level-load-monitoring-in-off-the-shelf-energy-monitors-a-technical-review.md b/content/publications/prospects-of-appliance-level-load-monitoring-in-off-the-shelf-energy-monitors-a-technical-review.md new file mode 100644 index 0000000..e8fb4d1 --- /dev/null +++ b/content/publications/prospects-of-appliance-level-load-monitoring-in-off-the-shelf-energy-monitors-a-technical-review.md @@ -0,0 +1,10 @@ ++++ +title = "Prospects of Appliance-Level Load Monitoring in Off-the-Shelf Energy Monitors: A Technical Review" +year = 2018 +authors = ["Anwar Ul Haq", "Hans-Arno Jacobsen"] +venue = "Energies" +publication_type = "Journal Article" +research = [] +external_url = "https://doi.org/10.3390/en11010189" +abstract = "The smart grid initiative has encouraged utility companies worldwide to roll-out new and smarter versions of energy meters. Before an extensive roll-out, which is both labor-intensive and incurs high capital costs, consumers need to be incentivised to reap the long-term benefits of such smart meters. Off-the-shelf energy monitors (e-monitors) can provide consumers with an insight into such potential benefits. As e-monitors are owned by the consumer, the consumer has greater control over the data, which significantly reduces the privacy and data confidentiality concerns. Because only limited online technical information is available about e-monitors, we evaluate several existing e-monitors using an online technical survey directly from the vendors. Besides automated e-monitoring, the use of different off-the-shelf e-monitors can also help to demonstrate state-of-the-art techniques such as non-intrusive load monitoring (NILM), data analytics, and the predictive maintenance of appliances. Our survey indicates a trend towards the incorporation of such state-of-the-art capabilities, particularly the appliance-level e-monitoring and load disaggregation. We have also discussed some essential requirements to implement load disaggregation in the next generation e-monitors. In future, these intelligent e-monitoring techniques will encourage effective consumer participation in the demand-side management (DSM) programs." ++++ diff --git a/content/publications/pta-a-programmable-teaching-assistant-for-lab-courses.md b/content/publications/pta-a-programmable-teaching-assistant-for-lab-courses.md index c87a4fe..b9e64dd 100644 --- a/content/publications/pta-a-programmable-teaching-assistant-for-lab-courses.md +++ b/content/publications/pta-a-programmable-teaching-assistant-for-lab-courses.md @@ -2,10 +2,10 @@ title = "pTA: A Programmable Teaching Assistant for Lab Courses" year = 2023 authors = ["Jawad Tahir", "Raj Mandal", "Olha Stefanova", "Hans-Arno Jacobsen", "Christoph Doblander", "Ruben Mayer"] -venue = "Proceedings of the 2nd International Workshop on Data Systems Education" +venue = "Proceedings of the 2nd International Workshop on Data Systems Education: Bridging education practice with education research" publication_type = "Conference Paper" research = ["distributed-machine-learning"] tags = ["education-systems", "lab-automation"] -summary = "Workshop paper on a programmable teaching assistant for systems-focused lab courses." -external_url = "https://dl.acm.org/doi/abs/10.1145/3596673.3596975" +external_url = "https://doi.org/10.1145/3596673.3596975" +abstract = "Lab courses play a crucial role in enabling students to gain a deeper understanding of theoretical concepts, but these courses require a significant effort from the course's organizational staff, such as instructors and teaching assistants. To address this challenge, we developed pTA, an acronym for programmable teaching assistant, which automates the functional evaluation of students' submissions for the Cloud Databases course taught at the Technical University of Munich (TUM). pTA reduces the staff workload and provides instant feedback to students, thereby enhancing their understanding of the project specifications. Additionally, pTA includes a live leaderboard that provides a gamification element that makes the course more interactive and engaging for students. It is deployed on a Kubernetes cluster that ensures scalability with evaluation requests. In this paper, we describe the course's learning milestones and provide an overview of pTA's architecture and features. The system's efficacy was evaluated at TUM and the University of Toronto, where it was deployed in two similar courses. Our findings show that pTA reduced the staff workload by at least 75%, lowered the operating cost, and increased course capacity in terms of the number of students. Furthermore, our study suggests that students exhibit more interest in courses that integrate interactive learning systems and gamification elements." +++ diff --git a/content/publications/publish-subscribe-for-mobile-applications-using-shared-dictionary-compression.md b/content/publications/publish-subscribe-for-mobile-applications-using-shared-dictionary-compression.md new file mode 100644 index 0000000..2df6f08 --- /dev/null +++ b/content/publications/publish-subscribe-for-mobile-applications-using-shared-dictionary-compression.md @@ -0,0 +1,10 @@ ++++ +title = "Publish/Subscribe for Mobile Applications Using Shared Dictionary Compression" +year = 2016 +authors = ["Christoph Doblander", "Kaiwen Zhang", "Hans-Arno Jacobsen"] +venue = "2016 IEEE 36th International Conference on Distributed Computing Systems (ICDCS)" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icdcs.2016.70" +abstract = "Publish/Subscribe is known as a scalable and efficient data dissemination mechanism. In a mobile environment, there is an added challenge for the pub/sub system to economizemobile bandwidth, which is especially precious in areas not wellcovered by mobile providers. While well-known compressionmethods such as GZip or Deflate are generally useful in suchsituations, we propose using Shared Dictionary Compression(SDC) to achieve a greater level of bandwidth efficiency. SDCrequires a dictionary, generated upfront, to be shared betweentwo communicating peers before it can be used. We proposea design where brokers forming the pub/sub overlay can be incharge of generating and propagating the shared dictionary. Oursolution employs an adaptive algorithm, executed at the brokers, which creates and maintains the dictionaries over time. Withthis approach, it is possible to reduce the required bandwidth byup to 88% including the introduced dictionary overhead. Ourdemo shows this approach applied to a smartphone applicationcommunicating with a publish/subscribe broker using the MQTTprotocol." ++++ diff --git a/content/publications/publish-subscribe-network-designs-for-multiplayer-games.md b/content/publications/publish-subscribe-network-designs-for-multiplayer-games.md new file mode 100644 index 0000000..de6d610 --- /dev/null +++ b/content/publications/publish-subscribe-network-designs-for-multiplayer-games.md @@ -0,0 +1,10 @@ ++++ +title = "Publish/subscribe network designs for multiplayer games" +year = 2014 +authors = ["César Cañas", "Kaiwen Zhang", "Bettina Kemme", "Jörg Kienzle", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 15th International Middleware Conference on - Middleware '14" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/2663165.2663337" +abstract = "Massively multiplayer online games (MMOGs), which are typically supported by large distributed systems, require a scalable, low latency messaging middleware that supports the location-based semantics and the loosely coupled interaction of multiplayer games components. In this paper, we present three different pub/sub-driven designs for a MMOG networking engine that account for the highly interactive and massive nature of these games. Each design uses not only different pub/sub approaches (from topic-based to content-based) but also serves varying degrees of responsibilities. In particular, some of them integrate game functionality, such as interest management, into the network engine. We implement, evaluate, and compare our proposed designs in the MMOG prototype Mammoth. Our real-world results show the viability of pub/sub while at the same time highlighting clear trade-offs between the different designs used, especially in the number and frequency of the various message types, such as subscriptions." ++++ diff --git a/content/publications/publisher-mobility-in-distributed-publish-subscribe-systems.md b/content/publications/publisher-mobility-in-distributed-publish-subscribe-systems.md new file mode 100644 index 0000000..d97b4a0 --- /dev/null +++ b/content/publications/publisher-mobility-in-distributed-publish-subscribe-systems.md @@ -0,0 +1,10 @@ ++++ +title = "Publisher Mobility in Distributed Publish/Subscribe Systems" +year = 2005 +authors = ["Vinod Muthusamy", "Milenko Petrovic", "Dapeng Gao", "Hans-Arno Jacobsen"] +venue = "25th IEEE International Conference on Distributed Computing Systems Workshops" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icdcsw.2005.116" +abstract = "The decoupling of producers and consumers in the publish/subscribe paradigm lends itself well to the support of mobile users who roam about the environment with intermittent network connectivity. This paper presents the first quantitative evaluation of publisher mobility in a distributed publish/subscribe system. Our results indicate that publisher mobility breaks a fundamental assumption of publish/subscribe systems and has a significant performance impact. We formalize publisher mobility algorithms for a distributed publish/subscribe system, and develop and evaluate optimizations to the mobile publisher algorithms." ++++ diff --git a/content/publications/publisher-placement-algorithms-in-content-based-publish-subscribe.md b/content/publications/publisher-placement-algorithms-in-content-based-publish-subscribe.md new file mode 100644 index 0000000..fcfe040 --- /dev/null +++ b/content/publications/publisher-placement-algorithms-in-content-based-publish-subscribe.md @@ -0,0 +1,10 @@ ++++ +title = "Publisher Placement Algorithms in Content-Based Publish/Subscribe" +year = 2010 +authors = ["Alex King Yeung Cheung", "Hans-Arno Jacobsen"] +venue = "2010 IEEE 30th International Conference on Distributed Computing Systems" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icdcs.2010.86" +abstract = "Many publish/subscribe systems implement a policy for clients to join to their physically closest broker to minimize transmission delays incurred on the clients' messages. However, the amount of delay reduced by this policy is only the tip of the iceberg as messages incur queuing, matching, transmission, and scheduling delays from traveling across potentially long distances in the broker network. Additionally, the clients' impact on system load is totally neglected by such a policy. This paper proposes two new algorithms that intelligently relocate publishers on the broker overlay to minimize both the overall end-to-end delivery delay and system load. Both algorithms exploit live publication distribution patterns but with different optimization metrics and computation methodologies to determine the best relocation point. Evaluations on PlanetLab and a cluster testbed show that our algorithms can reduce the average input load of the system by up to 68%, average broker message rate by up to 85%, and average delivery delay by up to 68%." ++++ diff --git a/content/publications/publiy-a-peer-assisted-publish-subscribe-service-for-timely-dissemination-of-bulk-content.md b/content/publications/publiy-a-peer-assisted-publish-subscribe-service-for-timely-dissemination-of-bulk-content.md new file mode 100644 index 0000000..00c1a43 --- /dev/null +++ b/content/publications/publiy-a-peer-assisted-publish-subscribe-service-for-timely-dissemination-of-bulk-content.md @@ -0,0 +1,10 @@ ++++ +title = "Publiy+: A Peer-Assisted Publish/Subscribe Service for Timely Dissemination of Bulk Content" +year = 2012 +authors = ["Reza Sherafat Kazemzadeh", "Hans-Arno Jacobsen"] +venue = "2012 IEEE 32nd International Conference on Distributed Computing Systems" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icdcs.2012.62" +abstract = "Publish/Subscribe (P/S) systems and file sharing applications traditionally share the common goal of disseminating data among large populations of users. Despite this similarity, the former focuses on timely dissemination of small-sized notification messages, while the latter presumes larger types of bulk content with less emphasis on the time needed between release and delivery of data. In this paper, we develop a peer-assisted content dissemination mechanism to bridge this gap by adopting the P/S model. We propose a hybrid two-layer architecture in which P/S brokers act as coordinators and guide their clients with interest in similar content to engage in direct exchange of data blocks in a peer-to-peer and cooperative fashion. Furthermore, we use network coding in order to facilitate data exchange among clients. Our peer-assisted scheme offloads the burden of disseminating huge volumes of data from P/S brokers to subscribers themselves. As an added advantage of our approach, brokers employ strategies that help shape traffic flows in multi-domain network settings. Finally, we have implemented our approach and carried out extensive large-scale experimental evaluation on a cluster with aggregate data transfers of up to 1 TB and involving up to 1000 subscribers. Our results demonstrate good scalability and faster content delivery compared to file sharing protocols such as BitTorrent." ++++ diff --git a/content/publications/q-dice-quantum-distributed-interconnect-compiler-and-emulator.md b/content/publications/q-dice-quantum-distributed-interconnect-compiler-and-emulator.md index 496cc0b..9838d35 100644 --- a/content/publications/q-dice-quantum-distributed-interconnect-compiler-and-emulator.md +++ b/content/publications/q-dice-quantum-distributed-interconnect-compiler-and-emulator.md @@ -7,5 +7,5 @@ publication_type = "ArXiv Preprint" research = ["quantum-computing-systems"] tags = ["quantum-systems"] external_url = "https://arxiv.org/abs/2606.11340" -source_url = "https://arxiv.org/abs/2606.11340" +abstract = "As distributed quantum computing (DQC) offers a leading path towards scalable quantum computation, the ability to benchmark distributed algorithms under realistic conditions becomes critical for system co-design. However, without access to physical systems, researchers lack tools to evaluate distribution protocols. We introduce Q-DICE (Quantum Distributed Interconnect Compiler and Emulator), a hardware-aware emulation environment for benchmarking distributed quantum circuits on classical simulators and on NISQ-era monolithic hardware. This work provides three core contributions: (1) a programmatic scheme to construct distributed QPU backends, utilizing two novel techniques - QPU slicing and stitching - to facilitate distributed circuit mapping, (2) a methodology for modeling nonlocal link noise using physically motivated Kraus operators and stochastic error channels, and (3) a boundary-aware circuit mapping algorithm enforcing distributed QPU topology constraints during transpilation. Together, these components constitute a distribution-aware compiler and noise-modeling engine that faithfully enforces the physical limitations of distributed quantum hardware within existing execution environments. We validate Q-DICE against a multitude of experimentally demonstrated quantum circuits, including a distributed Grover's search on optically linked trapped-ion hardware, achieving a worst-case fidelity deviation of 4% between simulated and experimental results. These findings demonstrate Q-DICE's capacity to accurately reproduce real distributed quantum system behavior across platforms, streamlining experimentation with distributed quantum algorithms and architectures." +++ diff --git a/content/publications/quantifying-aspects-in-middleware-platforms.md b/content/publications/quantifying-aspects-in-middleware-platforms.md new file mode 100644 index 0000000..e8a6e8a --- /dev/null +++ b/content/publications/quantifying-aspects-in-middleware-platforms.md @@ -0,0 +1,10 @@ ++++ +title = "Quantifying aspects in middleware platforms" +year = 2003 +authors = ["Charles Zhang", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 2nd international conference on Aspect-oriented software development" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/643603.643617" +abstract = "Middleware technologies such as Web Services, CORBA and DCOM have been very successful in solving distributed computing problems for a large family of application domains. As middleware systems are getting widely adopted and more functionally mature, it is also increasingly difficult for the architecture of middleware to achieve a high level of adaptability and configurability, due to the limitations of traditional software decomposition methods. Aspect oriented programming has brought us new design perspectives because it permits the superimpositions of multiple abstraction models on top of one another. It is a very powerful technique in separating and simplifying design concerns. In this paper, we first show that, through the quantification of aspects in the legacy implementations, the modularity of middleware architecture is greatly hindered by the ubiquitous existence of tangled logic. We then go one step further by factoring out a number of aspects identified in the mining work and re-implementing them as aspect programs. The aspect oriented re-factorization allows us to apply a set of software engineering metrics to quantify the changes of the re-factored system in both the structural complexity and the runtime performance. The aspect oriented re-factoring proves that the aspect oriented programming is capable of composing orthogonal design requirements. The final \"woven\" system is able to correctly provide both the fundamental functionality and the \"aspectized\" functionality with negligible overhead and a leaner architecture. Further more, the configurability of middleware is dramatically increased because the \"aspectized\" features can be configured in and out during the compile-time" ++++ diff --git a/content/publications/quantum-hypergraph-partitioning.md b/content/publications/quantum-hypergraph-partitioning.md new file mode 100644 index 0000000..ce6a803 --- /dev/null +++ b/content/publications/quantum-hypergraph-partitioning.md @@ -0,0 +1,10 @@ ++++ +title = "Quantum Hypergraph Partitioning" +year = 2026 +authors = ["Yiran Li", "Y. Batuhan Yilmaz", "Michael Silver", "Zachary Vernec", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 3rd workshop on Quantum Computing and Quantum-Inspired Technology for Data-Intensive Systems and Applications" +publication_type = "Conference Paper" +research = ["quantum-computing-systems", "data-management"] +external_url = "https://doi.org/10.1145/3811628.3811836" +abstract = "Hypergraph partitioning is a fundamental optimization problem with applications in data management and other domains involving higher-order relations. In this paper, we study balanced hypergraph partitioning from the perspective of quantum optimization. We formalize balanced k-way hypergraph partitioning with general hyperedge cut functions, and derive corresponding binary optimization formulations targeted at quantum optimization methods in both the two-way and multi-way settings. Our discussion highlights which cut functions admit Quadratic Unconstrained Binary Optimization (QUBO) encodings and which instead lead to higher-order binary objectives or rational forms. As a preliminary empirical validation, we focus on balanced two-way partitioning with the all-or-nothing cut on 3-uniform hypergraphs, where a direct QUBO is available, and evaluate simulated Quantum Approximate Optimization Algorithm (QAOA) and Simulated Annealing (SA) on small instances against exact solutions. The results show that the formulation is effective on small hypergraphs and that the balance-penalty weight plays a critical role in trading off cut quality and balance." ++++ diff --git a/content/publications/query-centric-partitioning-and-allocation-for-partially-replicated-database-systems.md b/content/publications/query-centric-partitioning-and-allocation-for-partially-replicated-database-systems.md new file mode 100644 index 0000000..d10a26b --- /dev/null +++ b/content/publications/query-centric-partitioning-and-allocation-for-partially-replicated-database-systems.md @@ -0,0 +1,10 @@ ++++ +title = "Query Centric Partitioning and Allocation for Partially Replicated Database Systems" +year = 2017 +authors = ["Tilmann Rabl", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 2017 ACM International Conference on Management of Data" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/3035918.3064052" +abstract = "A key feature of database systems is to provide transparent access to stored data. In distributed database systems, this includes data allocation and fragmentation. Transparent access introduces data dependencies and increases system complexity and inter-process communication. Therefore, many developers are exchanging transparency for better scalability using sharding and similar techniques. However, explicitly managing data distribution and data flow requires a deep understanding of the distributed system and the data access, and it reduces the possibilities for optimizations." ++++ diff --git a/content/publications/rapid-development-of-data-generators-using-meta-generators-in-pdgf.md b/content/publications/rapid-development-of-data-generators-using-meta-generators-in-pdgf.md new file mode 100644 index 0000000..32dba82 --- /dev/null +++ b/content/publications/rapid-development-of-data-generators-using-meta-generators-in-pdgf.md @@ -0,0 +1,10 @@ ++++ +title = "Rapid development of data generators using meta generators in PDGF" +year = 2013 +authors = ["Tilmann Rabl", "Meikel Poess", "Manuel Danisch", "Hans-Arno Jacobsen"] +venue = "Proceedings of the Sixth International Workshop on Testing Database Systems" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/2479440.2479441" +abstract = "Generating data sets for the performance testing of database systems on a particular hardware configuration and application domain is a very time consuming and tedious process. It is time consuming, because of the large amount of data that needs to be generated and tedious, because new data generators might need to be developed or existing once adjusted. The difficulty in generating this data is amplified by constant advances in hardware and software that allow the testing of ever larger and more complicated systems. In this paper, we present an approach for rapidly developing customized data generators. Our approach, which is based on the Parallel Data Generator Framework (PDGF), deploys a new concept of so called meta generators. Meta generators extend the concept of column-based generators in PDGF. Deploying meta generators in PDGF significantly reduces the development effort of customized data generators, it facilitates their debugging and eases their maintenance." ++++ diff --git a/content/publications/re-factoring-middleware-systems-a-case-study.md b/content/publications/re-factoring-middleware-systems-a-case-study.md new file mode 100644 index 0000000..73a4143 --- /dev/null +++ b/content/publications/re-factoring-middleware-systems-a-case-study.md @@ -0,0 +1,9 @@ ++++ +title = "Re-factoring Middleware Systems: A Case Study" +year = 2003 +authors = ["Charles Zhang", "Hans-Arno Jacobsen"] +venue = "Lecture Notes in Computer Science; On The Move to Meaningful Internet Systems 2003: CoopIS, DOA, and ODBASE" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1007/978-3-540-39964-3_79" ++++ diff --git a/content/publications/reaching-consensus-in-the-byzantine-empire.md b/content/publications/reaching-consensus-in-the-byzantine-empire.md index 2a5d701..dbab320 100644 --- a/content/publications/reaching-consensus-in-the-byzantine-empire.md +++ b/content/publications/reaching-consensus-in-the-byzantine-empire.md @@ -1,16 +1,11 @@ +++ title = "Reaching Consensus in the Byzantine Empire: A Comprehensive Review of BFT Consensus Algorithms" -year = 2023 +year = 2024 authors = ["Gengrui Zhang", "Fei Pan", "Yunhao Mao", "Sofia Tijanic", "Michael Dang'ana", "Shashank Motepalli", "Shiquan Zhang", "Hans-Arno Jacobsen"] venue = "ACM Computing Surveys" publication_type = "Journal Article" research = ["data-management"] tags = ["blockchain", "consensus", "systems-survey"] -summary = "Comprehensive review of Byzantine fault-tolerant consensus algorithms and their design tradeoffs." -external_url = "https://dl.acm.org/doi/abs/10.1145/3636553" +external_url = "https://doi.org/10.1145/3636553" +abstract = "Byzantine fault-tolerant (BFT) consensus algorithms are at the core of providing safety and liveness guarantees for distributed systems that must operate in the presence of arbitrary failures. Recently, numerous new BFT algorithms have been proposed, not least due to the traction blockchain technologies have garnered in the search for consensus solutions that offer high throughput, low latency, and robust system designs. In this article, we conduct a systematic survey of selected and distinguished BFT algorithms that have received extensive attention in academia and industry alike. We perform a qualitative comparison among all algorithms we review considering message and time complexities. Furthermore, we provide a comprehensive, step-by-step description of each surveyed algorithm by decomposing them into constituent subprotocols with intuitive figures to illustrate the message-passing pattern. We also elaborate on the strengths and weaknesses of each algorithm compared to the other state-of-the-art approaches." +++ - -This survey reviews prominent Byzantine fault-tolerant consensus algorithms and -compares them in terms of communication structure, time complexity, and overall -design tradeoffs. It also breaks each algorithm into constituent subprotocols -to make the design space easier to reason about. diff --git a/content/publications/real-time-load-prediction-with-high-velocity-smart-home-data-stream.md b/content/publications/real-time-load-prediction-with-high-velocity-smart-home-data-stream.md new file mode 100644 index 0000000..49b385d --- /dev/null +++ b/content/publications/real-time-load-prediction-with-high-velocity-smart-home-data-stream.md @@ -0,0 +1,10 @@ ++++ +title = "Real-time Load Prediction with High Velocity Smart Home Data Stream" +year = 2017 +authors = ["Christoph Doblander", "Martin Strohbach", "Holger Ziekow", "Hans-Arno Jacobsen"] +venue = "arXiv" +publication_type = "ArXiv Preprint" +research = ["data-management"] +external_url = "https://arxiv.org/abs/1708.04613" +abstract = "This paper addresses the use of smart-home sensor streams for continuous prediction of energy loads of individual households which participate as an agent in local markets. We introduces a new device level energy consumption dataset recorded over three years wich includes high resolution energy measurements from electrical devices collected within a pilot program. Using data from that pilot, we analyze the applicability of various machine learning mechanisms for continuous load prediction. Specifically, we address short-term load prediction that is required for load balancing in electrical micro-grids. We report on the prediction performance and the computational requirements of a broad range of prediction mechanisms. Furthermore we present an architecture and experimental evaluation when this prediction is applied in the stream." ++++ diff --git a/content/publications/reducing-communication-requirements-for-electric-vehicle-charging-using-vehicle-originating-signals.md b/content/publications/reducing-communication-requirements-for-electric-vehicle-charging-using-vehicle-originating-signals.md new file mode 100644 index 0000000..b1e6413 --- /dev/null +++ b/content/publications/reducing-communication-requirements-for-electric-vehicle-charging-using-vehicle-originating-signals.md @@ -0,0 +1,10 @@ ++++ +title = "Reducing communication requirements for electric vehicle charging using vehicle-originating-signals" +year = 2014 +authors = ["Victor del Razo", "Christoph Goebel", "Hans-Arno Jacobsen"] +venue = "2014 IEEE International Conference on Smart Grid Communications (SmartGridComm)" +publication_type = "Conference Paper" +research = [] +external_url = "https://doi.org/10.1109/smartgridcomm.2014.7007614" +abstract = "We propose a method for reducing the communication requirements for electric vehicle (EV) charging control based on Vehicle-Originating-Signals (VOS). The original VOS approach requires a message exchange between every EV and the aggregator for every time step. This can cause the load on communication systems to increase, especially as the number of EVs increases and the control intervals become more granular. We explore (i) the reduction of EV-originating messages, (ii) the reduction of aggregator-originating messages, and (iii) the challenges in combining both reduction methods. EVs reduce the number of messages by including possible future values in a single message. The aggregator reduces the number of messages by sending single broadcast signals. Combining both methods requires a retransmission protocol. For the evaluation, we compare the original and the improved VOS approaches on a load leveling scenario based on electricity demand, solar generation, and car mobility data from Munich, Germany. The results show that itis possible to reduce the overall message requirements with a minor effect on performance. With savings of over 70% in number of messages, these improvements contribute towards more network-friendly solutions for smart EV charging." ++++ diff --git a/content/publications/refactoring-middleware-with-aspects.md b/content/publications/refactoring-middleware-with-aspects.md new file mode 100644 index 0000000..16704c5 --- /dev/null +++ b/content/publications/refactoring-middleware-with-aspects.md @@ -0,0 +1,10 @@ ++++ +title = "Refactoring Middleware with Aspects" +year = 2003 +authors = ["Charles Zhang", "Hans-Arno Jacobsen"] +venue = "IEEE Transactions on Parallel and Distributed Systems" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.1109/tpds.2003.1247668" +abstract = "Middleware platforms, such as Web services, J2EE, CORBA, and DCOM, have become increasingly popular during the last decade. They have been very successful in solving distributed computing problems for a large family of application domains. The architecture of middleware systems have gone through many significant cycles of evolution, both in terms of the completeness of functionality and the range of adoptions for different types of platforms. However, at the same time, it is getting increasingly difficult to achieve and to maintain a high level of adaptability and configurability because the structure of the middleware architecture is becoming overly complicated and rigid. We attribute that problem to the limitations of traditional software decomposition methods. Aspect-oriented programming, on the contrary, has introduced new design perspectives that permit the superimpositions of different abstraction models on top of one another. This is a very powerful technique for separating and simplifying design concerns. In our effort of applying principles of aspect orientation to the middleware architecture, we first pragmatically analyze the use of aspects in the middleware architecture. We then show that aspects are the correct remedy for the above outlined middleware problems by quantifying crosscutting concerns in the legacy implementations of several prominent middleware systems. Our aspect analysis results strongly indicate that modularity of middleware architecture is greatly hindered by the wide existence of tangled logic. To go one step further, we factor out a number of crosscutting concerns identified in the mining process, reimplement them as aspects, and superimpose them back into the refactored architecture. This allows us to use a set of software engineering metrics to quantify the refactorization in terms of changes in the structural complexity, modularity, and performance of the resulting system. This aspect-oriented refactoring proves that aspect orientation is capable of composing orthogonal design requirements. The final \"woven\" system is able to correctly provide both the fundamental functionality and the \"aspectized\" functionality with negligible overhead and an overall leaner architecture. Furthermore, the \"aspectized\" feature can be configured in and out during compile-time, which greatly enhances the configurability of the architecture." ++++ diff --git a/content/publications/relevance-matters-capitalizing-on-less-top-k-matching-in-publish-subscribe.md b/content/publications/relevance-matters-capitalizing-on-less-top-k-matching-in-publish-subscribe.md new file mode 100644 index 0000000..9a67dff --- /dev/null +++ b/content/publications/relevance-matters-capitalizing-on-less-top-k-matching-in-publish-subscribe.md @@ -0,0 +1,10 @@ ++++ +title = "Relevance Matters: Capitalizing on Less (Top-k Matching in Publish/Subscribe)" +year = 2012 +authors = ["Mohammad Sadoghi", "Hans-Arno Jacobsen"] +venue = "2012 IEEE 28th International Conference on Data Engineering" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icde.2012.38" +abstract = "The efficient processing of large collections of Boolean expressions plays a central role in major data intensive applications ranging from user-centric processing and personalization to real-time data analysis. Emerging applications such as computational advertising and selective information dissemination demand determining and presenting to an end-user only the most relevant content that is both user-consumable and suitable for limited screen real estate of target devices. To retrieve the most relevant content, we present BE\\*-Tree, a novel indexing data structure designed for effective hierarchical top-k pattern matching, which as its by-product also reduces the operational cost of processing millions of patterns. To further reduce processing cost, BE\\*-Tree employs an adaptive and non-rigid space-cutting technique designed to efficiently index Boolean expressions over a high-dimensional continuous space. At the core of BE\\*-Tree lie two innovative ideas: (1) a bi-directional tree expansion build as a top-down (data and space clustering) and a bottom-up growths (space clustering), which together enable indexing only non-empty continuous sub-spaces, and (2) an overlap-free splitting strategy. Finally, the performance of BE\\*-Tree is proven through a comprehensive experimental comparison against state-of-the-art index structures for matching Boolean expressions." ++++ diff --git a/content/publications/reliable-and-highly-available-distributed-publish-subscribe-service.md b/content/publications/reliable-and-highly-available-distributed-publish-subscribe-service.md new file mode 100644 index 0000000..5823ac1 --- /dev/null +++ b/content/publications/reliable-and-highly-available-distributed-publish-subscribe-service.md @@ -0,0 +1,10 @@ ++++ +title = "Reliable and Highly Available Distributed Publish/Subscribe Service" +year = 2009 +authors = ["Reza Sherafat Kazemzadeh", "Hans-Arno Jacobsen"] +venue = "2009 28th IEEE International Symposium on Reliable Distributed Systems" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/srds.2009.32" +abstract = "This paper develops reliable distributed publish/subscriber algorithms with service availability in the face of concurrent crash failure of up to delta brokers. The reliability of service in our context refers to per-source in-order and exactly-once delivery of publications to matching subscribers. To handle failures, brokers maintain data structures that enable them to reconnect the topology and compute new forwarding paths on the fly. This enables fast reaction to failures and improves the system's availability. Moreover, we present a recovery procedure that recovering brokers execute in order to re-enter the system, and synchronize their routing information." ++++ diff --git a/content/publications/remon-remote-external-memory-over-the-network.md b/content/publications/remon-remote-external-memory-over-the-network.md new file mode 100644 index 0000000..533c669 --- /dev/null +++ b/content/publications/remon-remote-external-memory-over-the-network.md @@ -0,0 +1,9 @@ ++++ +title = "REMON: Remote External Memory Over the Network" +year = 2026 +authors = ["Shiquan Zhang", "Michail Bachras", "Yuqiu Zhang", "Yunhao Mao", "Hans-Arno Jacobsen"] +venue = "2026 IEEE 42nd International Conference on Data Engineering (ICDE)" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icde65706.2026.00175" ++++ diff --git a/content/publications/remop-remote-memory-aware-operator-optimization.md b/content/publications/remop-remote-memory-aware-operator-optimization.md new file mode 100644 index 0000000..6447fc7 --- /dev/null +++ b/content/publications/remop-remote-memory-aware-operator-optimization.md @@ -0,0 +1,10 @@ ++++ +title = "REMOP: REmote-Memory-aware OPerator Optimization" +year = 2026 +authors = ["Shiquan Zhang", "Yunhao Mao", "Yuqiu Zhang", "Gengrui Zhang", "Jeyhun Karimov", "Hans-Arno Jacobsen"] +venue = "arXiv" +publication_type = "ArXiv Preprint" +research = ["data-management"] +external_url = "https://arxiv.org/abs/2606.19576" +abstract = "Remote and disaggregated memory tiers expand the effective memory capacity of analytical database engines, but they also reshape the cost structure of out-of-memory query processing. When an operator spills beyond local DRAM, moving pages to remote memory incurs both data-transfer time and a fixed round-trip latency per transfer. Classical operator analyses and buffer-allocation heuristics primarily target disk spilling by minimizing total I/O volume. Under remote memory, these strategies can be suboptimal because they may trigger excessive transfer rounds. We present REMOP, a remote-memory-aware operator optimization framework that uses transfer-round-aware intra-operator memory policies to improve out-of-memory execution under tight memory budgets. REMOP introduces the number of transfer rounds into the latency cost model and derives operator-specific buffer-partitioning strategies, instantiating the approach for blocked nested-loop join, external merge sort, and external hash join in DuckDB. Our evaluation on a two-node compute-memory testbed shows that REMOP reduces transfer rounds by up to 97% and operator runtime by up to 48% on spill-heavy microbenchmarks, and lowers the average runtime of spilling TPC-H and TPC-DS queries by 22.7% and 26.4% end-to-end." ++++ diff --git a/content/publications/representation-learning-for-appliance-recognition-a-comparison-to-classical-machine-learning.md b/content/publications/representation-learning-for-appliance-recognition-a-comparison-to-classical-machine-learning.md new file mode 100644 index 0000000..0e5b519 --- /dev/null +++ b/content/publications/representation-learning-for-appliance-recognition-a-comparison-to-classical-machine-learning.md @@ -0,0 +1,10 @@ ++++ +title = "Representation Learning for Appliance Recognition: A Comparison to Classical Machine Learning" +year = 2022 +authors = ["Matthias Kahl", "Daniel Jorde", "Hans-Arno Jacobsen"] +venue = "arXiv" +publication_type = "ArXiv Preprint" +research = ["distributed-machine-learning"] +external_url = "https://arxiv.org/abs/2209.03759" +abstract = "Non-intrusive load monitoring (NILM) aims at energy consumption and appliance state information retrieval from aggregated consumption measurements, with the help of signal processing and machine learning algorithms. Representation learning with deep neural networks is successfully applied to several related disciplines. The main advantage of representation learning lies in replacing an expert-driven, hand-crafted feature extraction with hierarchical learning from many representations in raw data format. In this paper, we show how the NILM processing-chain can be improved, reduced in complexity and alternatively designed with recent deep learning algorithms. On the basis of an event-based appliance recognition approach, we evaluate seven different classification models: a classical machine learning approach that is based on a hand-crafted feature extraction, three different deep neural network architectures for automated feature extraction on raw waveform data, as well as three baseline approaches for raw data processing. We evaluate all approaches on two large-scale energy consumption datasets with more than 50,000 events of 44 appliances. We show that with the use of deep learning, we are able to reach and surpass the performance of the state-of-the-art classical machine learning approach for appliance recognition with an F-Score of 0.75 and 0.86 compared to 0.69 and 0.87 of the classical approach." ++++ diff --git a/content/publications/resolving-feature-convolution-in-middleware-systems.md b/content/publications/resolving-feature-convolution-in-middleware-systems.md new file mode 100644 index 0000000..ccc9455 --- /dev/null +++ b/content/publications/resolving-feature-convolution-in-middleware-systems.md @@ -0,0 +1,10 @@ ++++ +title = "Resolving feature convolution in middleware systems" +year = 2004 +authors = ["Charles Zhang", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 19th annual ACM SIGPLAN conference on Object-oriented programming, systems, languages, and applications" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/1028976.1028992" +abstract = "Middleware provides simplicity and uniformity for the development of distributed applications. However, the modularity of the architecture of middleware is starting to disintegrate and to become complicated due to the interaction of too many orthogonal concerns imposed from a wide range of application requirements. This is not due to bad design but rather due to the limitations of the conventional architectural decomposition methodologies. We introduce the principles of horizontal decomposition (HD) which addresses this problem with a mixed-paradigm middleware architecture. HD provides guidance for the use of conventional decomposition methods to implement the core functionalities of middleware and the use of aspect orientation to address its orthogonal properties. Our evaluation of the horizontal decomposition principles focuses on refactoring major middleware functionalities into aspects in order to modularize and isolate them from the core architecture. New versions of the middleware platform can be created through combining the core and the flexible selection of middleware aspects such as IDL data types, the oneway invocation style, the dynamic messaging style, and additional character encoding schemes. As a result, the primary functionality of the middleware is supported with a much simpler architecture and enhanced performance. Moreover, customization and configuration of the middleware for a wide-range of requirements becomes possible." ++++ diff --git a/content/publications/retrofitting-admission-control-in-an-internet-scale-application.md b/content/publications/retrofitting-admission-control-in-an-internet-scale-application.md new file mode 100644 index 0000000..993c86f --- /dev/null +++ b/content/publications/retrofitting-admission-control-in-an-internet-scale-application.md @@ -0,0 +1,10 @@ ++++ +title = "Retrofitting Admission Control in an Internet-Scale Application" +year = 2016 +authors = ["Tanmay Chaudhry", "Christoph Doblander", "Anatol Dammer", "Cristian Klein", "Hans-Arno Jacobsen"] +venue = "Umeå University, UMINF 16.17" +publication_type = "Technical Report" +research = [] +external_url = "https://webapps.cs.umu.se/uminf/reports/2016/017/part1.pdf" +abstract = "In this paper we propose a methodology to retrofit admission control in an Internet-scale, production application. Admission control requires less effort to improve the availability of an application, in particular when making it scalable is costly. This can occur due to the integration of 3rd-party legacy code or handling large amounts of data, and is further motivated by lean thinking, which argues for building a minimum viable product to discover customer requirements. Our main contribution consists in a method to generate an amplified workload, that is realistic enough to test all kinds of what-if scenarios, but does not require an exhaustive transition matrix. This workload generator can then be used to iteratively stress-test the application, identify the next bottleneck and add admission control. To illustrate the usefulness of the approach, we report on our experience with adding admission control within SimScale, a Software-as-a-Service start-up for engineering simulations, that already features 50,000 users." ++++ diff --git a/content/publications/reversible-conflict-free-replicated-data-types.md b/content/publications/reversible-conflict-free-replicated-data-types.md index 60c5dbb..5a201f0 100644 --- a/content/publications/reversible-conflict-free-replicated-data-types.md +++ b/content/publications/reversible-conflict-free-replicated-data-types.md @@ -6,6 +6,6 @@ venue = "Proceedings of the 23rd ACM/IFIP International Middleware Conference" publication_type = "Conference Paper" research = ["data-management"] tags = ["distributed-systems"] -external_url = "https://dl.acm.org/doi/10.1145/3528535.3565252" -source_url = "https://msrg.utoronto.ca/publications/?page=2" +external_url = "https://doi.org/10.1145/3528535.3565252" +abstract = "Conflict-free replicated data types (CRDTs) are popular for optimistic replication and ensuring strong eventual consistency (SEC) in distributed systems. However, reversibility is an underdeveloped functionality for CRDTs, despite its usefulness in system restoration from an erroneous state or undoing unwanted operations. In this paper, we define the concept and design of reversible CRDTs (rCRDTs). Reverse operations compensate for the effect of reversed updates, and they extend existing CRDT interfaces. Three abstractions for reversibility are proposed: reversing a single update, multiple causally related updates, and multiple logically related updates that capture the user intention behind the updates. Moreover, a replicated and distributed key-value store, rKVCRDT, is implemented as a proof of concept that integrates the support of reversible CRDTs. The rCRDTs' evaluation show that although adding reversibility affects the system's performance, the end result depends on multiple factors and varies based on the underlying CRDTs. System designers must consider the trade-off between the benefit of reversibility and the performance impact." +++ diff --git a/content/publications/reward-mechanism-for-blockchains-using-evolutionary-game-theory.md b/content/publications/reward-mechanism-for-blockchains-using-evolutionary-game-theory.md index c58c4ae..d711305 100644 --- a/content/publications/reward-mechanism-for-blockchains-using-evolutionary-game-theory.md +++ b/content/publications/reward-mechanism-for-blockchains-using-evolutionary-game-theory.md @@ -6,6 +6,6 @@ venue = "2021 3rd Conference on Blockchain Research & Applications for Innovativ publication_type = "Conference Paper" research = ["data-management"] tags = ["blockchain"] -external_url = "https://ieeexplore.ieee.org/document/9569791/" -source_url = "https://msrg.utoronto.ca/publications/?page=2" +external_url = "https://doi.org/10.1109/brains52497.2021.9569791" +abstract = "Blockchains have witnessed widespread adoption in the past decade in various fields. The growing demand makes their scalability and sustainability challenges more evident than ever. As a result, more and more blockchains have begun to adopt proof-of-stake (PoS) consensus protocols to address those challenges. One of the fundamental characteristics of any blockchain technology is its crypto-economics and incentives. Lately, each PoS blockchain has designed a unique reward mechanism, yet, many of them are prone to free-rider and nothing-at-stake problems. To better understand the ad-hoc design of reward mechanisms, in this paper, we develop a reward mechanism framework that could apply to many PoS blockchains. We formulate the block validation game wherein the rewards are distributed for validating the blocks correctly. Using evolutionary game theory, we analyze how the participants' behaviour could potentially evolve with the reward mechanism. Also, penalties are found to play a central role in maintaining the integrity of blockchains." +++ diff --git a/content/publications/routing-of-xml-and-xpath-queries-in-data-dissemination-networks.md b/content/publications/routing-of-xml-and-xpath-queries-in-data-dissemination-networks.md new file mode 100644 index 0000000..cc5a3cc --- /dev/null +++ b/content/publications/routing-of-xml-and-xpath-queries-in-data-dissemination-networks.md @@ -0,0 +1,10 @@ ++++ +title = "Routing of XML and XPath Queries in Data Dissemination Networks" +year = 2008 +authors = ["Guoli Li", "Shuang Hou", "Hans-Arno Jacobsen"] +venue = "2008 The 28th International Conference on Distributed Computing Systems" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icdcs.2008.31" +abstract = "XML-based data dissemination networks are rapidly gaining momentum. In these networks XML content is routed from data producers to data consumers throughout an overlay network of content-based routers. Routing decisions are based on XPath expressions (XPEs) stored at each router. To enable efficient routing, while keeping the routing state small, we introduce an advertisement-based routing algorithm for XML content, present a novel data structure for managing XPEs, especially apt for the hierarchical nature of XPEs and XML, and develop several optimizations for reducing the number of XPEs required to manage the routing state. The experimental evaluation shows that our algorithms and optimizations reduce the routing table size by up to 90%, improve the routing time by roughly 85%, and reduce overall network traffic by about 35%. Experiments running on PlanetLab show the scalability of our approach." ++++ diff --git a/content/publications/s-topss-semantic-toronto-publish-subscribe-system.md b/content/publications/s-topss-semantic-toronto-publish-subscribe-system.md new file mode 100644 index 0000000..b35a44d --- /dev/null +++ b/content/publications/s-topss-semantic-toronto-publish-subscribe-system.md @@ -0,0 +1,10 @@ ++++ +title = "S-ToPSS: Semantic Toronto Publish/Subscribe System" +year = 2003 +authors = ["Milenko Petrovic", "Ioana Burcea", "Hans-Arno Jacobsen"] +venue = "Proceedings 2003 VLDB Conference" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1016/b978-012722442-8/50115-4" +abstract = "The increase in the amount of data on the Internet has led to the development of a new generation of applications based on selective information dissemination where, data is distributed only to interested clients. Such applications require a new middleware architecture that can efficiently match user interests with available information. Middleware that can satisfy this requirement include event-based architectures such as publish-subscribe systems. In this demonstration paper we address the problem of semantic matching. We investigate how current publish/subscribe systems can be extended with semantic capabilities. Our main contribution is the development and validation (through demonstration) of a semantic pub/sub system prototype S-ToPSS (Semantic Toronto Publish/Subscribe System)." ++++ diff --git a/content/publications/safe-distribution-and-parallel-execution-of-data-centric-workflows-over-the-publish-subscribe-abstraction.md b/content/publications/safe-distribution-and-parallel-execution-of-data-centric-workflows-over-the-publish-subscribe-abstraction.md new file mode 100644 index 0000000..459786e --- /dev/null +++ b/content/publications/safe-distribution-and-parallel-execution-of-data-centric-workflows-over-the-publish-subscribe-abstraction.md @@ -0,0 +1,10 @@ ++++ +title = "Safe Distribution and Parallel Execution of Data-Centric Workflows over the Publish/Subscribe Abstraction" +year = 2015 +authors = ["Mohammad Sadoghi", "Martin Jergler", "Hans-Arno Jacobsen", "Richard Hull", "Roman Vaculín"] +venue = "IEEE Transactions on Knowledge and Data Engineering" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.1109/tkde.2015.2421331" +abstract = "In this work, we develop an approach for the safe distribution and parallel execution of data-centric workflows over the publish/subscribe abstraction. In essence, we design a unique representation of data-centric workflows, specifically designed to exploit the loosely coupled and distributed nature of publish/subscribe systems. Furthermore, we argue for the practicality and expressiveness of our approach by mapping a standard and industry-strength data-centric workflow model, namely, IBM Business Artifacts with Guard-Stage-Milestone (GSM), into the publish/subscribe abstraction. In short, the contributions of this work are three-fold: (1) mapping of data-centric workflows into publish/subscribe to achieve distributed and parallel execution; (2) detailed theoretical analysis of the mapping; and (3) formulation of the complexity of the optimal workflow distribution over the publish/subscribe abstraction as an NP-hard problem." ++++ diff --git a/content/publications/scalable-deep-learning-on-distributed-infrastructures-challenges-techniques-and-tools.md b/content/publications/scalable-deep-learning-on-distributed-infrastructures-challenges-techniques-and-tools.md new file mode 100644 index 0000000..b594d93 --- /dev/null +++ b/content/publications/scalable-deep-learning-on-distributed-infrastructures-challenges-techniques-and-tools.md @@ -0,0 +1,10 @@ ++++ +title = "Scalable Deep Learning on Distributed Infrastructures: Challenges, Techniques, and Tools" +year = 2021 +authors = ["Ruben Mayer", "Hans-Arno Jacobsen"] +venue = "ACM Computing Surveys" +publication_type = "Journal Article" +research = ["distributed-machine-learning"] +external_url = "https://doi.org/10.1145/3363554" +abstract = "Deep Learning (DL) has had an immense success in the recent past, leading to state-of-the-art results in various domains, such as image recognition and natural language processing. One of the reasons for this success is the increasing size of DL models and the proliferation of vast amounts of training data being available. To keep on improving the performance of DL, increasing the scalability of DL systems is necessary. In this survey, we perform a broad and thorough investigation on challenges, techniques and tools for scalable DL on distributed infrastructures. This incorporates infrastructures for DL, methods for parallel DL training, multi-tenant resource scheduling, and the management of training and model data. Further, we analyze and compare 11 current open-source DL frameworks and tools and investigate which of the techniques are commonly implemented in practice. Finally, we highlight future research trends in DL systems that deserve further research." ++++ diff --git a/content/publications/scalable-multiway-stream-joins-in-hardware.md b/content/publications/scalable-multiway-stream-joins-in-hardware.md new file mode 100644 index 0000000..5b4f3e8 --- /dev/null +++ b/content/publications/scalable-multiway-stream-joins-in-hardware.md @@ -0,0 +1,10 @@ ++++ +title = "Scalable Multiway Stream Joins in Hardware" +year = 2020 +authors = ["Mohammadreza Najafi", "Mohammad Sadoghi", "Hans-Arno Jacobsen"] +venue = "IEEE Transactions on Knowledge and Data Engineering" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.1109/tkde.2019.2916860" +abstract = "Efficient real-time analytics are an integral part of an increasing number of data management applications, such as computational targeted advertising, algorithmic trading, and Internet of Things. In this paper, we focus primarily on accelerating stream joins, which are arguably one of the most commonly used and resource-intensive operators in stream processing. We propose a scalable circular pipeline design (Circular-MJ) in hardware to orchestrate a multiway join while minimizing data flow disruption. In this circular design, each new tuple (given its origin stream) starts its processing from a specific join core and passes through all respective join cores in a pipeline sequence to produce the final results. We also present a novel two-stage pipeline stream join (Stashed-MJ) that uses a best-effort buffering technique (referred to as stash) to maintain intermediate results. If an overwrite is detected in the stash, our design automatically resorts to recomputing intermediate results. Finally, we present a parallelized version of our multiway stream join by integrating our proposed pipelines into a parallel unidirectional flow-based architecture (Parallel-MJ). Our experimental results demonstrate a linear throughput scaling with respect to the numbers of streams and processing cores." ++++ diff --git a/content/publications/scaling-construction-of-low-fan-out-overlays-for-topic-based-publish-subscribe-systems.md b/content/publications/scaling-construction-of-low-fan-out-overlays-for-topic-based-publish-subscribe-systems.md new file mode 100644 index 0000000..ec9e0f7 --- /dev/null +++ b/content/publications/scaling-construction-of-low-fan-out-overlays-for-topic-based-publish-subscribe-systems.md @@ -0,0 +1,10 @@ ++++ +title = "Scaling Construction of Low Fan-out Overlays for Topic-Based Publish/Subscribe Systems" +year = 2011 +authors = ["Chen Chen", "Roman Vitenberg", "Hans-Arno Jacobsen"] +venue = "2011 31st International Conference on Distributed Computing Systems" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icdcs.2011.68" +abstract = "It is a key challenge and fundamental problem in the design of distributed publish/subscribe systems to construct the underlying dissemination overlay. In this paper, we focus on effective practical solution for the Min Max-TCO problem: Create a topic-connected pub/sub overlay in which all nodes interested in the same topic are organized in a directly connected dissemination sub-overlay while keeping the maximum node degree to the minimum. Previously known solutions provided an extensive analysis of the problem and an algorithm that achieves a logarithmic approximation for Min Max-TCO. Yet, they did not focus on efficiency of the solution or feasibility of decentralized operation that would not require full knowledge of the system. Compared to these solutions, our proposed algorithm produces an overlay with marginally higher degrees. At the same time, it has drastically reduced runtime cost, which is corroborated by both theoretical analysis and empirical evaluation. The latter shows a speedup by a factor of more than 25 on average for typical pub/sub workloads." ++++ diff --git a/content/publications/scientific-workflow-mining-in-clouds.md b/content/publications/scientific-workflow-mining-in-clouds.md new file mode 100644 index 0000000..7573018 --- /dev/null +++ b/content/publications/scientific-workflow-mining-in-clouds.md @@ -0,0 +1,10 @@ ++++ +title = "Scientific Workflow Mining in Clouds" +year = 2017 +authors = ["Wei Song", "Fangfei Chen", "Hans-Arno Jacobsen", "Xiaoxu Xia", "Chunyang Ye", "Xiaoxing Ma"] +venue = "IEEE Transactions on Parallel and Distributed Systems" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.1109/tpds.2017.2696942" +abstract = "Computing clouds have become the platform of choice for the deployment and execution of scientific workflows. Due to the uncertainty and unpredictability of scientific exploration, the execution plan for a scientific workflow may vary from the definition. It is therefore of great significance to be able to discover actual workflows from execution histories (event logs) to reproduce experimental results and to establish provenance. However, most existing process mining techniques focus on discovering control flow-oriented business processes in a centralized environment, and thus, they are mostly inapplicable to the discovery of data flow-oriented, unstructured scientific workflows in distributed cloud environments. In this paper, we present Scientific Workflow Mining as a Service (SWMaaS) to support both intra-cloud and inter-cloud scientific workflow mining. The approach is implemented as a ProM plug-in and is evaluated on event logs derived from real-world scientific workflows. Through experimental results, we demonstrate the effectiveness and efficiency of our approach." ++++ diff --git a/content/publications/scope-fl-a-strategy-proof-chain-based-optimal-pareto-efficient-federated-learning-system.md b/content/publications/scope-fl-a-strategy-proof-chain-based-optimal-pareto-efficient-federated-learning-system.md new file mode 100644 index 0000000..c5cd7a1 --- /dev/null +++ b/content/publications/scope-fl-a-strategy-proof-chain-based-optimal-pareto-efficient-federated-learning-system.md @@ -0,0 +1,10 @@ ++++ +title = "SCOPE-FL: A Strategy-proof Chain-based Optimal pareto efficient Federated Learning System" +year = 2026 +authors = ["Seyed Salar Ghazi", "Kaiwen Zhang", "Mehdi feizi", "Hans-Arno Jacobsen"] +venue = "arXiv" +publication_type = "ArXiv Preprint" +research = ["distributed-machine-learning"] +external_url = "https://arxiv.org/abs/2606.18384" +abstract = "Hierarchical Federated Learning (HFL) enables scalable collaborative model training across distributed devices while preserving data privacy. However, existing HFL client selection mechanisms suffer from a fundamental strategic inefficiency. By prioritizing stability over Pareto efficiency (PE), they produce suboptimal resource allocations, and without strategy proofness (SP), participants are incentivized to misrepresent their true preferences, both failures degrading system overall welfare in the Pareto sense in practice. To address it, we propose SCOPE-FL (Strategy-proof Chain-based Optimal pareto efficient Federated Learning), a synchronous HFL framework that formulates client selection as a two-sided school choice problem solved through the Top Trading Cycle (TTC) algorithm that simultaneously guarantees PE and SP. For reward distribution, SCOPE-FL employs a scalable Shapley value approximation based on One-Round Reconstruction (OR), ensuring compensation proportional to each client's contribution. The entire mechanism executes via blockchain smart contracts, providing the tamper-proof environment required for the SP guarantees to hold in practice. A comprehensive evaluation on MNIST, Fashion-MNIST, and CIFAR-10 demonstrates that SCOPE-FL outperforms state-of-the-art approaches, including DA, IAS, and other methods across model accuracy, convergence rate, and reward efficiency, while achieving communication latency comparable to DA and blockchain overhead significantly lower than DA at scale." ++++ diff --git a/content/publications/sdn-like-the-next-generation-of-pub-sub.md b/content/publications/sdn-like-the-next-generation-of-pub-sub.md new file mode 100644 index 0000000..dbf9886 --- /dev/null +++ b/content/publications/sdn-like-the-next-generation-of-pub-sub.md @@ -0,0 +1,10 @@ ++++ +title = "SDN-like: The Next Generation of Pub/Sub" +year = 2013 +authors = ["Kaiwen Zhang", "Hans-Arno Jacobsen"] +venue = "arXiv" +publication_type = "ArXiv Preprint" +research = ["data-management"] +external_url = "https://arxiv.org/abs/1308.0056" +abstract = "Software-Defined Networking (SDN) has raised the boundaries of cloud computing by offering unparalleled levels of control and flexibility to system administrators over their virtualized environments. To properly embrace this new era of SDN-driven network architectures, the research community must not only consider the impact of SDN over the protocol stack, but also on its overlying networked applications. In this big ideas paper, we study the impact of SDN on the design of future message-oriented middleware, specifically pub/sub systems. We argue that key concepts put forth by SDN can be applied in a meaningful fashion to the next generation of pub/sub systems. First, pub/sub can adopt a logically centralized controller model for maintenance, monitoring, and control of the overlay network. We establish a parallel with existing work on centralized pub/sub routing and discuss how the logically centralized controller model can be implemented in a distributed manner. Second, we investigate the separation of the control and data plane, which is integral to SDN, which can be adopted to raise the level of decoupling in pub/sub. We introduce a new model of pub/sub which separates the traditional publisher and subscriber roles into flow regulators and producer/consumers of data. We then present use cases that benefit from this approach and study the impact of decoupling for performance." ++++ diff --git a/content/publications/self-evolving-subscriptions-for-content-based-publish-subscribe-systems.md b/content/publications/self-evolving-subscriptions-for-content-based-publish-subscribe-systems.md new file mode 100644 index 0000000..86c752c --- /dev/null +++ b/content/publications/self-evolving-subscriptions-for-content-based-publish-subscribe-systems.md @@ -0,0 +1,10 @@ ++++ +title = "Self-Evolving Subscriptions for Content-Based Publish/Subscribe Systems" +year = 2017 +authors = ["César Cañas", "Kaiwen Zhang", "Bettina Kemme", "Jörg Kienzle", "Hans-Arno Jacobsen"] +venue = "2017 IEEE 37th International Conference on Distributed Computing Systems (ICDCS)" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icdcs.2017.277" +abstract = "Traditional pub/sub systems cannot adequately handle workloads of applications with dynamic, short-lived subscriptions such as location-based social networks, predictive stock trading, and online games. Subscribers must continuously interact with the pub/sub system to remove and insert subscriptions, thereby inefficiently consuming network and computing resources, and sacrificing consistency. In the aforementioned applications, we recognize that the changes in the subscriptions can follow a predictable pattern over some variable (e.g., time). In this paper, we present a new type of subscription, called evolving subscription, which encapsulates these patterns and allow the pub/sub system to autonomously adapt to the dynamic interests of the subscribers without incurring an expensive re-subscription overhead. We propose a general model for expressing evolving subscriptions and a framework for supporting them in a pub/sub system. To this end, we propose three different designs to support evolving subscriptions, which are evaluated and compared to the traditional resubscription approach in the context of two use cases: online games and high-frequency trading. Our evaluation shows that our solutions can reduce subscription traffic by 96.8% and improve delivery accuracy when compared to the baseline resubscription mechanism." ++++ diff --git a/content/publications/server-provisioning-in-content-delivery-clouds.md b/content/publications/server-provisioning-in-content-delivery-clouds.md new file mode 100644 index 0000000..3ee8042 --- /dev/null +++ b/content/publications/server-provisioning-in-content-delivery-clouds.md @@ -0,0 +1,10 @@ ++++ +title = "Server Provisioning in Content Delivery Clouds" +year = 2015 +authors = ["Kianoosh Mokhtarian", "Hans-Arno Jacobsen"] +venue = "2015 IEEE 8th International Conference on Cloud Computing" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/cloud.2015.19" +abstract = "Cloud based video delivery platforms serve a significant fraction of the entire Internet traffic and are continuously expanding with the growing demand. We study the provisioning of the server cluster to deploy at each location in such systems. We optimize the right server count, peering bandwidth, and server configuration as the disk size and the necessary SSD and/or RAM caches to sustain the intensive I/O load. Our analyses are based on actual server traces from a global content delivery platform. Our optimization captures the interaction of cache layers in each server, the interplay between egress/disk capacity and network bandwidth, storage read/write constraints and storage prices." ++++ diff --git a/content/publications/service-subscription-and-consumption-for-personal-web-applications.md b/content/publications/service-subscription-and-consumption-for-personal-web-applications.md new file mode 100644 index 0000000..e8411c6 --- /dev/null +++ b/content/publications/service-subscription-and-consumption-for-personal-web-applications.md @@ -0,0 +1,9 @@ ++++ +title = "Service Subscription and Consumption for Personal Web Applications" +year = 2013 +authors = ["Chunyang Ye", "Young Yoon", "Hans-Arno Jacobsen"] +venue = "Lecture Notes in Computer Science; The Personal Web" +publication_type = "Book Chapter" +research = [] +external_url = "https://doi.org/10.1007/978-3-642-39995-4_3" ++++ diff --git a/content/publications/shared-dictionary-compression-in-publish-subscribe-systems.md b/content/publications/shared-dictionary-compression-in-publish-subscribe-systems.md new file mode 100644 index 0000000..8416416 --- /dev/null +++ b/content/publications/shared-dictionary-compression-in-publish-subscribe-systems.md @@ -0,0 +1,10 @@ ++++ +title = "Shared dictionary compression in publish/subscribe systems" +year = 2016 +authors = ["Christoph Doblander", "Tanuj Ghinaiya", "Kaiwen Zhang", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 10th ACM International Conference on Distributed and Event-based Systems" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/2933267.2933308" +abstract = "Publish/subscribe is known as a scalable and efficient data dissemination mechanism. Its efficiency comes from the optimized routing algorithms, yet few works exist on employing compression to save bandwidth, which is especially important in mobile environments. State of the art compression methods such as GZip or Deflate can be generally employed to compress messages. In this paper, we show how to reduce bandwidth even further by employing Shared Dictionary Compression (SDC) in pub/sub. However, SDC requires a dictionary to be generated and disseminated prior to compression, which introduces additional computational and bandwidth overhead. To support SDC, we propose a novel and lightweight protocol for pub/sub which employs a new class of brokers, called sampling brokers. Our solution generates, and disseminates dictionaries using the sampling brokers. Dictionary maintenance is performed regularly using an adaptive algorithm. The evaluation of our proposed design shows that it is possible to compensate for the introduced overhead and achieve significant bandwidth reduction over Deflate." ++++ diff --git a/content/publications/should-my-blockchain-learn-to-drive-a-study-of-hyperledger-fabric.md b/content/publications/should-my-blockchain-learn-to-drive-a-study-of-hyperledger-fabric.md new file mode 100644 index 0000000..4a575d5 --- /dev/null +++ b/content/publications/should-my-blockchain-learn-to-drive-a-study-of-hyperledger-fabric.md @@ -0,0 +1,10 @@ ++++ +title = "Should my Blockchain Learn to Drive? A Study of Hyperledger Fabric" +year = 2024 +authors = ["Jeeta Ann Chacko", "Ruben Mayer", "Hans-Arno Jacobsen"] +venue = "arXiv" +publication_type = "ArXiv Preprint" +research = ["data-management"] +external_url = "https://arxiv.org/abs/2406.06318" +abstract = "Similar to other transaction processing frameworks, blockchain systems need to be dynamically reconfigured to adapt to varying workloads and changes in network conditions. However, achieving optimal reconfiguration is particularly challenging due to the complexity of the blockchain stack, which has diverse configurable parameters. This paper explores the concept of self-driving blockchains, which have the potential to predict workload changes and reconfigure themselves for optimal performance without human intervention. We compare and contrast our discussions with existing research on databases and highlight aspects unique to blockchains. We identify specific parameters and components in Hyperledger Fabric, a popular permissioned blockchain system, that are suitable for autonomous adaptation and offer potential solutions for the challenges involved. Further, we implement three demonstrative locally autonomous systems, each targeting a different layer of the blockchain stack, and conduct experiments to understand the feasibility of our findings. Our experiments indicate up to 11% improvement in success throughput and a 30% decrease in latency, making this a significant step towards implementing a fully autonomous blockchain system in the future." ++++ diff --git a/content/publications/size-does-not-matter-sparsification-and-graph-neural-network-sampling-for-large-scale-graphs.md b/content/publications/size-does-not-matter-sparsification-and-graph-neural-network-sampling-for-large-scale-graphs.md new file mode 100644 index 0000000..e375f15 --- /dev/null +++ b/content/publications/size-does-not-matter-sparsification-and-graph-neural-network-sampling-for-large-scale-graphs.md @@ -0,0 +1,11 @@ ++++ +title = "Size Does (Not) Matter? Sparsification and Graph Neural Network Sampling for Large-scale Graphs" +year = 2024 +authors = ["Jana Vatter", "Maurice L. Rochau", "Ruben Mayer", "Hans-Arno Jacobsen"] +venue = "VLDB Workshops" +publication_type = "Conference Paper" +research = ["distributed-machine-learning"] +external_url = "https://vldb.org/workshops/2024/proceedings/LSGDA/LSGDA24.06.pdf" +abstract = "With the ever-growing size of real-world graphs, optimizing Graph Neural Network (GNN) training has become essential. Two key methods for this are graph sparsification and GNN sampling, both aim at reducing graph size while preserving valuable information. The question arises what kind of information should be preserved and what a reasonable graph size is. We propose combining random graph sparsification with GNN sampling, showing that this approach can significantly reduce training time while maintaining accuracy. Our experiments demonstrate that sparsification to around 40% of the original graph and sampling with a fanout parameter of 4 yields the best results in terms of training time and accuracy. Beyond training time, also inference time can be decreased up to 75% which enables scalability for time-critical applications such as fraud detection. Finally, we identify open challenges and new research directions, including sampling-aware graph reduction methods, mining new graph datasets, and the prevention of bias." +abstract_license_url = "https://creativecommons.org/licenses/by-nc-nd/4.0/" ++++ diff --git a/content/publications/sla-driven-business-process-management-in-soa-2009.md b/content/publications/sla-driven-business-process-management-in-soa-2009.md new file mode 100644 index 0000000..a7a342b --- /dev/null +++ b/content/publications/sla-driven-business-process-management-in-soa-2009.md @@ -0,0 +1,10 @@ ++++ +title = "SLA-driven business process management in SOA" +year = 2009 +authors = ["Vinod Muthusamy", "Hans-Arno Jacobsen", "Tony Chau", "Allen Chan", "Phil Coulthard"] +venue = "Proceedings of the 2009 Conference of the Center for Advanced Studies on Collaborative Research - CASCON '09" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/1723028.1723040" +abstract = "The management of non-functional goals, or Service Level Agreements (SLA), in the development of business processes in a Service Oriented Architecture (SOA) often requires much manual and error-prone effort by all parties throughout the entire lifecycle of the processes. The formal specification of SLAs into development tools can simplify some of this effort. In particular, the runtime provisioning and monitoring of processes can be achieved by an autonomic system that adapts to changing conditions to maintain the SLA's goals. SOA supports partitioning a system into services that are running in a distributed execution environment. When coupled with an associated cost model, a process can be both executed and monitored in an optimal manner, based on a declarative, user-specified optimality function." ++++ diff --git a/content/publications/sla-driven-business-process-management-in-soa.md b/content/publications/sla-driven-business-process-management-in-soa.md new file mode 100644 index 0000000..0cad236 --- /dev/null +++ b/content/publications/sla-driven-business-process-management-in-soa.md @@ -0,0 +1,10 @@ ++++ +title = "SLA-driven business process management in SOA" +year = 2007 +authors = ["Vinod Muthusamy", "Hans-Arno Jacobsen", "Phil Coulthard", "Allen Chan", "Julie Waterhouse", "Elena Litani"] +venue = "Proceedings of the 2007 conference of the center for advanced studies on Collaborative research - CASCON '07" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/1321211.1321243" +abstract = "In a Service-Oriented Architecture (SOA), distributed applications are built by orchestrating reusable services using high-level workflows or business processes. The complexity of developing and maintaining these processes is addressed by SOA development cycles that identify the roles of participants at each stage. To assist development, sophisticated tools have been developed, such as the IBM® WebSphere® suite of SOA products. However, the development, administration and maintenance of a business process still requires much manual effort that can be automated. In particular, the non-functional goals of a business process, often expressed as Service Level Agreements (SLA) need to be manually considered at each stage of the development process." ++++ diff --git a/content/publications/sla-driven-distributed-application-development.md b/content/publications/sla-driven-distributed-application-development.md new file mode 100644 index 0000000..918b8fb --- /dev/null +++ b/content/publications/sla-driven-distributed-application-development.md @@ -0,0 +1,10 @@ ++++ +title = "SLA-driven distributed application development" +year = 2008 +authors = ["Vinod Muthusamy", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 3rd workshop on Middleware for service oriented computing" +publication_type = "Conference Paper" +research = [] +external_url = "https://doi.org/10.1145/1462802.1462808" +abstract = "The management of Service Level Agreements (SLA) in the development of business processes in a Service Oriented Architecture (SOA) often requires much manual and errorprone effort by all parties throughout the lifecycle of the processes. The formal specification of SLAs into development tools can simplify some of this effort. In particular, the runtime provisioning and monitoring of processes can be achieved by an autonomic system that adapts to changing conditions to maintain the SLA's goals. A cost model allows the efficient execution and monitoring of processes, based on a declarative, user-specified optimality function. Experiments demonstrate that the system can indeed adapt to changing workload conditions, saving roughly 70% of the network bandwidth in one particular experiment." ++++ diff --git a/content/publications/small-scale-peer-to-peer-publish-subscribe.md b/content/publications/small-scale-peer-to-peer-publish-subscribe.md new file mode 100644 index 0000000..efc07af --- /dev/null +++ b/content/publications/small-scale-peer-to-peer-publish-subscribe.md @@ -0,0 +1,10 @@ ++++ +title = "Small Scale Peer-to-Peer Publish/Subscribe" +year = 2005 +authors = ["Vinod Muthusamy", "Hans-Arno Jacobsen"] +venue = "P2PKM" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://ceur-ws.org/Vol-139/2.pdf" +abstract = "The scalability of publish/subscribe (pub/sub) systems and distributed hash tables (DHTs) have been extensively studied in the literature. However, less well-known are properties of the pub/sub model and DHTs that make them suitable for small networks. This paper articulates these benefits, and evaluates the performance of a DHT-based pub/sub implementation in small-scale networks. We find that a fundamental assumption of DHT-based data management applications is violated in small networks, and this makes the pub/sub implementation exhibit poor load balance under certain workloads. This work illustrates that data management applications that scale in large networks may not scale in small networks." ++++ diff --git a/content/publications/smart-buildings-and-smart-grids-dagstuhl-seminar-15091.md b/content/publications/smart-buildings-and-smart-grids-dagstuhl-seminar-15091.md new file mode 100644 index 0000000..013787b --- /dev/null +++ b/content/publications/smart-buildings-and-smart-grids-dagstuhl-seminar-15091.md @@ -0,0 +1,10 @@ ++++ +title = "Smart Buildings and Smart Grids (Dagstuhl Seminar 15091)" +year = 2015 +authors = ["Hans-Arno Jacobsen", "Randy H. Katz", "Hartmut Schmeck", "Christoph Goebel"] +venue = "Dagstuhl Reports" +publication_type = "Journal Article" +research = [] +external_url = "https://doi.org/10.4230/dagrep.5.2.128" +abstract = "This report provides an overview of the program, discussions, and outcomes of Dagstuhl Seminar 15091 \"Smart Buildings and Smart Grids\", which took place from 22-27 February 2015 at Schloss Dagstuhl - Leibniz Center for Informatics. The main goal of the seminar was to provide a forum for leading Energy Informatics (EI) researchers to discuss their recent research on Smart Buildings and Smart Grids, to further elaborate EI research agenda and methods, and to kick-start new research projects with industry. The report contains abstracts of talks that were held by the participants and the outcomes of several discussion sessions on the focal topics of the seminar (e.g., information technology driven developments in building and power system management, as well as cross-cutting topics, such as computer networks, data management, and system design." ++++ diff --git a/content/publications/smart-charging-schedules-for-highway-travel-with-electric-vehicles.md b/content/publications/smart-charging-schedules-for-highway-travel-with-electric-vehicles.md new file mode 100644 index 0000000..600df52 --- /dev/null +++ b/content/publications/smart-charging-schedules-for-highway-travel-with-electric-vehicles.md @@ -0,0 +1,10 @@ ++++ +title = "Smart Charging Schedules for Highway Travel With Electric Vehicles" +year = 2016 +authors = ["Victor del Razo", "Hans-Arno Jacobsen"] +venue = "IEEE Transactions on Transportation Electrification" +publication_type = "Journal Article" +research = [] +external_url = "https://doi.org/10.1109/tte.2016.2560524" +abstract = "Electric vehicles (EVs) can contribute to reducing carbon emissions and facilitate renewable integration. However, EVs are not competitive with fuel-based vehicles, particularly for long distances, because of their limited range and long charging times. We propose a smart scheduling approach for EVs to plan charging stops on a highway with limited charging infrastructure. This approach aims to minimize the total travel time for each EV based on the A\\* algorithm with constraint verification and a peer-to-peer scheduling system. By considering the estimated state of the charging stations, we achieve indirect coordination between EVs. We introduce a simulation framework with trips generated using a data-driven approach and support for time-varying highway parameters. Furthermore, we apply our approach to a use case for the German highway A9 from Munich to Berlin. The computation and communication requirements of the proposed solution remain moderate and privacy preserving, contributing to its applicability. Results show that the smart scheduling approach significantly reduces the total travel times. In addition, by dynamically adjusting the schedules, the proposed approach can account for changing highway conditions, for example, slow traffic on a given segment. Our approach can be generalized beyond fast charging to different technologies, such as hydrogen or battery swapping stations." ++++ diff --git a/content/publications/solving-big-data-challenges-for-enterprise-application-performance-management.md b/content/publications/solving-big-data-challenges-for-enterprise-application-performance-management.md new file mode 100644 index 0000000..c5f6ab9 --- /dev/null +++ b/content/publications/solving-big-data-challenges-for-enterprise-application-performance-management.md @@ -0,0 +1,10 @@ ++++ +title = "Solving Big Data Challenges for Enterprise Application Performance Management" +year = 2012 +authors = ["Tilmann Rabl", "Mohammad Sadoghi", "Hans-Arno Jacobsen", "Sergio Gómez-Villamor", "Victor Muntés-Mulero", "Serge Mankowskii"] +venue = "Proceedings of the VLDB Endowment" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.14778/2367502.2367512" +abstract = "As the complexity of enterprise systems increases, the need for monitoring and analyzing such systems also grows. A number of companies have built sophisticated monitoring tools that go far beyond simple resource utilization reports. For example, based on instrumentation and specialized APIs, it is now possible to monitor single method invocations and trace individual transactions across geographically distributed systems. This high-level of detail enables more precise forms of analysis and prediction but comes at the price of high data rates (i.e., big data). To maximize the benefit of data monitoring, the data has to be stored for an extended period of time for ulterior analysis. This new wave of big data analytics imposes new challenges especially for the application performance monitoring systems. The monitoring data has to be stored in a system that can sustain the high data rates and at the same time enable an up-to-date view of the underlying infrastructure. With the advent of modern key-value stores, a variety of data storage systems have emerged that are built with a focus on scalability and high data rates as predominant in this monitoring use case. In this work, we present our experience and a comprehensive performance evaluation of six modern (open-source) data stores in the context of application performance monitoring as part of CA Technologies initiative. We evaluated these systems with data and workloads that can be found in application performance monitoring, as well as, on-line advertisement, power monitoring, and many other use cases. We present our insights not only as performance results but also as lessons learned and our experience relating to the setup and configuration complexity of these data stores in an industry setting." ++++ diff --git a/content/publications/solving-manufacturing-equipment-monitoring-through-efficient-complex-event-processing-debs-grand-challenge.md b/content/publications/solving-manufacturing-equipment-monitoring-through-efficient-complex-event-processing-debs-grand-challenge.md new file mode 100644 index 0000000..82724ea --- /dev/null +++ b/content/publications/solving-manufacturing-equipment-monitoring-through-efficient-complex-event-processing-debs-grand-challenge.md @@ -0,0 +1,10 @@ ++++ +title = "Solving manufacturing equipment monitoring through efficient complex event processing: DEBS grand challenge" +year = 2012 +authors = ["Tilmann Rabl", "Kaiwen Zhang", "Mohammad Sadoghi", "Navneet Kumar Pandey", "Aakash Nigam", "Chen Wang", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 6th ACM International Conference on Distributed Event-Based Systems" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/2335484.2335520" +abstract = "In this paper, we present an efficient complex event processing system tailored toward monitoring a large-scale setup of manufacturing equipment. In particular, the key challenge in the equipment monitoring is to develop an event-based system for computing complex manufacturing queries coupled with event notifications and event and query result visualization components. Furthermore, we present an experimental evaluation to validate the effectiveness of the proposed solution with respect to both query latency and throughput." ++++ diff --git a/content/publications/splitjoin-a-scalable-low-latency-stream-join-architecture-with-adjustable-ordering-precision.md b/content/publications/splitjoin-a-scalable-low-latency-stream-join-architecture-with-adjustable-ordering-precision.md new file mode 100644 index 0000000..bf313de --- /dev/null +++ b/content/publications/splitjoin-a-scalable-low-latency-stream-join-architecture-with-adjustable-ordering-precision.md @@ -0,0 +1,10 @@ ++++ +title = "SplitJoin: A Scalable, Low-latency Stream Join Architecture with Adjustable Ordering Precision" +year = 2016 +authors = ["Mohammadreza Najafi", "Mohammad Sadoghi", "Hans-Arno Jacobsen"] +venue = "USENIX ATC" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://www.usenix.org/conference/atc16/technical-sessions/presentation/najafi" +abstract = "There is a rising interest in accelerating stream processing through modern parallel hardware, yet it remains a challenge as how to exploit the available resources to achieve higher throughput without sacrificing latency due to the increased length of processing pipeline and communication path and the need for central coordination. To achieve these objectives, we introduce a novel top-down data flow model for stream join processing (arguably, one of the most resource-intensive operators in stream processing), called SplitJoin, that operates by splitting the join operation into independent storing and processing steps that gracefully scale with respect to the number of cores. Furthermore, SplitJoin eliminates the need for global coordination while preserving the order of input streams by re-thinking how streams are channeled into distributed join computation cores and maintaining the order of output streams by proposing a novel distributed punctuation technique. Throughout our experimental analysis, SplitJoin offered up to 60% improvement in throughput while reducing latency by up to 3.3X compared to state-of-the-art solutions." ++++ diff --git a/content/publications/sregym-a-live-benchmark-for-ai-sre-agents-with-high-fidelity-failure-scenarios.md b/content/publications/sregym-a-live-benchmark-for-ai-sre-agents-with-high-fidelity-failure-scenarios.md new file mode 100644 index 0000000..e953352 --- /dev/null +++ b/content/publications/sregym-a-live-benchmark-for-ai-sre-agents-with-high-fidelity-failure-scenarios.md @@ -0,0 +1,10 @@ ++++ +title = "SREGym: A Live Benchmark for AI SRE Agents with High-Fidelity Failure Scenarios" +year = 2026 +authors = ["Jackson Clark", "Yiming Su", "Saad Mohammad Rafid Pial", "Yifang Tian", "Lily Gniedziejko", "Hans-Arno Jacobsen", "Yinfang Chen", "Tianyin Xu"] +venue = "arXiv" +publication_type = "ArXiv Preprint" +research = ["distributed-machine-learning"] +external_url = "https://arxiv.org/abs/2605.07161" +abstract = "AI agents are increasingly used to diagnose and mitigate failures in production systems, known as agentic Site Reliability Engineering (SRE). Current SRE benchmarks are limited to oversimplistic SRE tasks and are unfortunately hard to extend due to bespoke designs. We present SREGym, a high-fidelity benchmark for SRE agents. SREGym exposes a live system environment built atop real-world cloud-native system stacks, where high-fidelity failure scenarios are simulated through fault injectors. SREGym models the complexity of production environments by simulating (1) a wide range of faults at different layers, (2) various ambient noises, and (3) diverse failure modes such as metastable failures and correlated failures. SREGym is architected as a modular, extensible framework that orchestrates fault and noise injectors across stacks. SREGym currently includes 90 realistic, challenging SRE problems. We use SREGym to evaluate frontier agents and show that their capabilities varies significantly in addressing different kinds of failures, with up to 40% differences in end-to-end results. SREGym is actively maintained as an open-source project and has been used by researchers and practitioners." ++++ diff --git a/content/publications/subscription-covering-for-relevance-based-filtering-in-content-based-publish-subscribe-systems.md b/content/publications/subscription-covering-for-relevance-based-filtering-in-content-based-publish-subscribe-systems.md new file mode 100644 index 0000000..0fe92cc --- /dev/null +++ b/content/publications/subscription-covering-for-relevance-based-filtering-in-content-based-publish-subscribe-systems.md @@ -0,0 +1,10 @@ ++++ +title = "Subscription Covering for Relevance-Based Filtering in Content-Based Publish/Subscribe Systems" +year = 2017 +authors = ["Kaiwen Zhang", "Vinod Muthusamy", "Mohammad Sadoghi", "Hans-Arno Jacobsen"] +venue = "2017 IEEE 37th International Conference on Distributed Computing Systems (ICDCS)" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icdcs.2017.184" +abstract = "Large-scale applications require a scalable data dissemination service with advanced filtering capabilities. We propose the use of a content-based publish/subscribe system with support for top-k filtering in the context of such applications. We focus on the problem of top-k subscription filtering, where a publication is delivered only to the k highest scoring subscribers. The naive approach to perform filtering early at the publisher edge works only if complete knowledge of the subscriptions is available, which is not compatible with the well-established covering optimization in scalable content-based publish/subscribe systems. We propose an efficient rank-cover technique to reconcile top-k subscription filtering with covering. We extend the covering model to support top-k and describe a novel algorithm for forwarding subscriptions to publishers while maintaining correctness. Finally, we compare our solutions to a baseline covering system. In a typical setting, our optimized solution is scalable and provides over 81% of the covering benefit." ++++ diff --git a/content/publications/the-debs-2021-grand-challenge-analyzing-environmental-impact-of-worldwide-lockdowns.md b/content/publications/the-debs-2021-grand-challenge-analyzing-environmental-impact-of-worldwide-lockdowns.md index 3504fc8..7abc28f 100644 --- a/content/publications/the-debs-2021-grand-challenge-analyzing-environmental-impact-of-worldwide-lockdowns.md +++ b/content/publications/the-debs-2021-grand-challenge-analyzing-environmental-impact-of-worldwide-lockdowns.md @@ -6,6 +6,6 @@ venue = "Proceedings of the 15th ACM International Conference on Distributed and publication_type = "Conference Paper" research = ["data-management"] tags = ["benchmarking"] -external_url = "https://dl.acm.org/doi/10.1145/3465480.3467836" -source_url = "https://msrg.utoronto.ca/publications/?page=2" +external_url = "https://doi.org/10.1145/3465480.3467836" +abstract = "The ACM DEBS 2021 Grand Challenge (GC) is the eleventh episode of a series of programming challenge competitions that began in 2011. Every year, participants of the GC are provided with new datasets and practical problems, and the challenge receives novel and high performant solutions from research, academia, and industry. The theme of the DEBS '21 GC is analyzing the environmental effects of COVID-19 restrictions. This year's edition of the GC is the first to explicitly focus on the integration and practicability of the solutions by fostering the use of distributed solutions based on widely-used open-source platforms and by requiring participants to address non-functional properties besides correctness of the solution. This paper describes the dataset used, formalizes the problem statement, and explains the evaluation platform that made dataset distribution and remote evaluation possible with our new virtualized infrastructure." +++ diff --git a/content/publications/the-debs-2022-grand-challenge-detecting-trading-trends-in-financial-tick-data.md b/content/publications/the-debs-2022-grand-challenge-detecting-trading-trends-in-financial-tick-data.md index 453fd2e..42197d3 100644 --- a/content/publications/the-debs-2022-grand-challenge-detecting-trading-trends-in-financial-tick-data.md +++ b/content/publications/the-debs-2022-grand-challenge-detecting-trading-trends-in-financial-tick-data.md @@ -1,5 +1,5 @@ +++ -title = "The DEBS 2022 Grand Challenge: Detecting Trading Trends in Financial Tick Data" +title = "Detecting trading trends in financial tick data: the DEBS 2022 grand challenge" year = 2022 related_datasets = ["debs-2022-trading-data"] authors = ["Sebastian Frischbier", "Jawad Tahir", "Christoph Doblander", "Arne Hormann", "Ruben Mayer", "Hans-Arno Jacobsen"] @@ -7,6 +7,6 @@ venue = "Proceedings of the 16th ACM International Conference on Distributed and publication_type = "Conference Paper" research = ["data-management"] tags = ["benchmarking", "event-processing", "stream-processing"] -summary = "DEBS Grand Challenge paper on detecting trading trends in financial tick data." -external_url = "https://arxiv.org/abs/2206.13237" +external_url = "https://doi.org/10.1145/3524860.3539645" +abstract = "The DEBS Grand Challenge (GC) is an annual programming competition open to practitioners from both academia and industry. The GC 2022 edition focuses on real-time complex event processing of high-volume tick data provided by Infront Financial Technology GmbH. The goal of the challenge is to efficiently compute specific trend indicators and detect patterns in these indicators like those used by real-life traders to decide on buying or selling in financial markets. The data set Trading Data used for benchmarking contains 289 million tick events from approximately 5500+ financial instruments that had been traded on the three major exchanges Amsterdam (NL), Paris (FR), and Frankfurt am Main (GER) over the course of a full week in 2021. The data set is made publicly available. In addition to correctness and performance, submissions must explicitly focus on reusability and practicability. Hence, participants must address specific nonfunctional requirements and are asked to build upon open-source platforms. This paper describes the required scenario and the data set Trading Data, defines the queries of the problem statement, and explains the enhancements made to the evaluation platform Challenger that handles data distribution, dynamic subscriptions, and remote evaluation of the submissions." +++ diff --git a/content/publications/the-debs-2025-grand-challenge-real-time-monitoring-of-defects-in-laser-powder-bed-fusion-l-pbf-manufacturing.md b/content/publications/the-debs-2025-grand-challenge-real-time-monitoring-of-defects-in-laser-powder-bed-fusion-l-pbf-manufacturing.md new file mode 100644 index 0000000..86df306 --- /dev/null +++ b/content/publications/the-debs-2025-grand-challenge-real-time-monitoring-of-defects-in-laser-powder-bed-fusion-l-pbf-manufacturing.md @@ -0,0 +1,9 @@ ++++ +title = "The DEBS 2025 Grand Challenge: Real-Time Monitoring of Defects in Laser Powder Bed Fusion (L-PBF) Manufacturing" +year = 2025 +authors = ["Luca De Martini", "Jawad Tahir", "Alessandro Margara", "Christoph Doblander", "Sebastian Frischbier", "Ruben Mayer", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 19th ACM International Conference on Distributed and Event-based Systems" +publication_type = "Conference Paper" +research = [] +external_url = "https://doi.org/10.1145/3701717.3735578" ++++ diff --git a/content/publications/the-evolution-of-distributed-systems-for-graph-neural-networks-and-their-origin-in-graph-processing-and-deep-learning-a-survey.md b/content/publications/the-evolution-of-distributed-systems-for-graph-neural-networks-and-their-origin-in-graph-processing-and-deep-learning-a-survey.md new file mode 100644 index 0000000..01b1483 --- /dev/null +++ b/content/publications/the-evolution-of-distributed-systems-for-graph-neural-networks-and-their-origin-in-graph-processing-and-deep-learning-a-survey.md @@ -0,0 +1,10 @@ ++++ +title = "The Evolution of Distributed Systems for Graph Neural Networks and Their Origin in Graph Processing and Deep Learning: A Survey" +year = 2024 +authors = ["Jana Vatter", "Ruben Mayer", "Hans-Arno Jacobsen"] +venue = "ACM Computing Surveys" +publication_type = "Journal Article" +research = ["distributed-machine-learning"] +external_url = "https://doi.org/10.1145/3597428" +abstract = "Graph neural networks (GNNs) are an emerging research field. This specialized deep neural network architecture is capable of processing graph structured data and bridges the gap between graph processing and deep learning. As graphs are everywhere, GNNs can be applied to various domains including recommendation systems, computer vision, natural language processing, biology, and chemistry. With the rapid growing size of real-world graphs, the need for efficient and scalable GNN training solutions has come. Consequently, many works proposing GNN systems have emerged throughout the past few years. However, there is an acute lack of overview, categorization, and comparison of such systems. We aim to fill this gap by summarizing and categorizing important methods and techniques for large-scale GNN solutions. Additionally, we establish connections between GNN systems, graph processing systems, and deep learning systems." ++++ diff --git a/content/publications/the-fqp-vision-flexible-query-processing-on-a-reconfigurable-computing-fabric.md b/content/publications/the-fqp-vision-flexible-query-processing-on-a-reconfigurable-computing-fabric.md new file mode 100644 index 0000000..46dfc40 --- /dev/null +++ b/content/publications/the-fqp-vision-flexible-query-processing-on-a-reconfigurable-computing-fabric.md @@ -0,0 +1,10 @@ ++++ +title = "The FQP Vision: Flexible Query Processing on a Reconfigurable Computing Fabric" +year = 2015 +authors = ["Mohammadreza Najafi", "Mohammad Sadoghi", "Hans-Arno Jacobsen"] +venue = "ACM SIGMOD Record" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.1145/2814710.2814712" +abstract = "The Flexible Query Processor (FQP) constitutes a family of hardware-based data stream processors that support dynamic changes to queries and streams, as well as static changes to the processor-internal fabric in order to maximize performance for given workloads. FQP is prototyped on field-programmable gate arrays (FPGAs). To this end, FQP supports select, project and window-join queries over data streams. While processing incoming tuples, FQP can accept new queries, a key characteristic distinguishing FQP from related approaches employing FPGAs for stream processing. In this paper, we present our vision of FQP, focusing on few internal details to support the flexibility dimension, in particular, the segment-at-a-time mechanism to realize processing of tuples of variable sizes. While many of these features are readily available in software, their hardware-based realizations have been one of the main shortcomings of existing research efforts" ++++ diff --git a/content/publications/the-impact-of-state-of-charge-management-when-providing-regulation-power-with-energy-storage.md b/content/publications/the-impact-of-state-of-charge-management-when-providing-regulation-power-with-energy-storage.md new file mode 100644 index 0000000..b261cf3 --- /dev/null +++ b/content/publications/the-impact-of-state-of-charge-management-when-providing-regulation-power-with-energy-storage.md @@ -0,0 +1,10 @@ ++++ +title = "The Impact of State of Charge Management When Providing Regulation Power With Energy Storage" +year = 2014 +authors = ["Christoph Goebel", "Duncan S. Callaway", "Hans-Arno Jacobsen"] +venue = "IEEE Transactions on Power Systems" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.1109/tpwrs.2013.2292434" +abstract = "Recent work suggests that further integration of wind and solar resources into the power grid will increase the demand for regulation and load following. The purpose of this letter is to investigate an important aspect of using energy storages for providing regulation: The short-term energy restoration of such limited energy resources (LERs) coincides with load following operations and could therefore lead to higher load following requirements. We therefore analyze the impact of corresponding scheduling approaches on load following energy and capacity using stochastic simulations. In particular, we compare the load following impact of a basic control strategy proposed by the stakeholders of the Independent System Operator in California (CAISO) to a strategy that attempts to manage the energy level of LERs such that their impact on load following requirements is optimal. Our results show that such smart state of charge management could even reduce the demand for load following while satisfying the full regulation demand in the CAISO control region." ++++ diff --git a/content/publications/the-padres-distributed-publish-subscribe-system.md b/content/publications/the-padres-distributed-publish-subscribe-system.md new file mode 100644 index 0000000..3bb735f --- /dev/null +++ b/content/publications/the-padres-distributed-publish-subscribe-system.md @@ -0,0 +1,9 @@ ++++ +title = "The PADRES Distributed Publish/Subscribe System" +year = 2005 +authors = ["E. Fidler", "Hans-Arno Jacobsen", "Guoli Li", "Serge Mankovski"] +venue = "FIW" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://dblp.org/rec/conf/fiw/FidlerJLM05" ++++ diff --git a/content/publications/the-padres-event-processing-network-uniform-querying-of-past-and-future-events-das-padres-ereignisverarbeitungsnetzwerk-einheitliche-anfragen-auf-ereignisse-der-vergangenheit-und-zukunft.md b/content/publications/the-padres-event-processing-network-uniform-querying-of-past-and-future-events-das-padres-ereignisverarbeitungsnetzwerk-einheitliche-anfragen-auf-ereignisse-der-vergangenheit-und-zukunft.md new file mode 100644 index 0000000..230b001 --- /dev/null +++ b/content/publications/the-padres-event-processing-network-uniform-querying-of-past-and-future-events-das-padres-ereignisverarbeitungsnetzwerk-einheitliche-anfragen-auf-ereignisse-der-vergangenheit-und-zukunft.md @@ -0,0 +1,10 @@ ++++ +title = "The PADRES Event Processing Network: Uniform Querying of Past and Future Events (Das PADRES Ereignisverarbeitungsnetzwerk: Einheitliche Anfragen auf Ereignisse der Vergangenheit und Zukunft)" +year = 2009 +authors = ["Hans-Arno Jacobsen", "Vinod Muthusamy", "Guoli Li"] +venue = "itit" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.1524/itit.2009.0549" +abstract = "This paper outlines requirements and sketches techniques for introducing to the publish/subscribe model the capability to uniformly access data produced in the past and future. The new model can filter, aggregate, correlate and project any combination of historic and future data. A flexible architecture is presented consisting of distributed and replicated data repositories that can be provisioned in ways to tradeoff availability, storage overhead, query overhead, query delay, load distribution, parallelism, redundancy and locality." ++++ diff --git a/content/publications/the-padres-publish-subscribe-system.md b/content/publications/the-padres-publish-subscribe-system.md new file mode 100644 index 0000000..7dfc381 --- /dev/null +++ b/content/publications/the-padres-publish-subscribe-system.md @@ -0,0 +1,10 @@ ++++ +title = "The PADRES Publish/Subscribe System" +year = 2010 +authors = ["Hans-Arno Jacobsen", "Alex King Yeung Cheung", "Guoli Li", "Balasubramaneyam Maniymaran", "Vinod Muthusamy", "Reza Sherafat Kazemzadeh"] +venue = "Principles and Applications of Distributed Event-Based Systems" +publication_type = "Book Chapter" +research = ["data-management"] +external_url = "https://doi.org/10.4018/9781605666976.ch008" +abstract = "This chapter introduces PADRES, the publish/subscribe model with the capability to correlate events, uniformly access data produced in the past and future, balance the traffic load among brokers, and handle network failures. The new model can filter, aggregate, correlate and project any combination of historic and future data. A flexible architecture is proposed consisting of distributed and replicated data repositories that can be provisioned in ways to tradeoff availability, storage overhead, query overhead, query delay, load distribution, parallelism, redundancy and locality. This chapter gives a detailed overview of the PADRES content-based publish/subscribe system. Several applications are presented in detail that can benefit from the content-based nature of the publish/subscribe paradigm and take advantage of its scalability and robustness features. A list of example applications are discussed that can benefit from the content-based nature of publish/subscribe paradigm and take advantage of its scalability and robustness features.Request access from your librarian to read this chapter's full text." ++++ diff --git a/content/publications/the-potential-of-smart-home-sensors-in-forecasting-household-electricity-demand.md b/content/publications/the-potential-of-smart-home-sensors-in-forecasting-household-electricity-demand.md new file mode 100644 index 0000000..5c13e58 --- /dev/null +++ b/content/publications/the-potential-of-smart-home-sensors-in-forecasting-household-electricity-demand.md @@ -0,0 +1,10 @@ ++++ +title = "The potential of smart home sensors in forecasting household electricity demand" +year = 2013 +authors = ["Holger Ziekow", "Christoph Goebel", "Jens Strüker", "Hans-Arno Jacobsen"] +venue = "2013 IEEE International Conference on Smart Grid Communications (SmartGridComm)" +publication_type = "Conference Paper" +research = [] +external_url = "https://doi.org/10.1109/smartgridcomm.2013.6687962" +abstract = "The aim of this paper is to quantify the impact of disaggregated electric power measurements on the accuracy of household demand forecasts. Demand forecasting on the household level is regarded as an essential mechanism for matching distributed power generation and demand in smart power grids. We use state-of-the-art forecasting tools, in particular support vector machines and neural networks, to evaluate the use of disaggregated smart home sensor data for household-level demand forecasting. Our investigation leverages high resolution data from 3 private households collected over 30 days. Our key results are as follows: First, by comparing the accuracy of the machine learning based forecasts with a persistence forecast we show that advanced forecasting methods already yield better forecasts, even when carried out on aggregated household consumption data that could be obtained from smart meters (1-7%). Second, our comparison of forecasts based on disaggregated data from smart home sensors with the persistence and smart meter benchmarks reveals further forecast improvements (4-33%). Third, our sensitivity analysis with respect to the time resolution of data shows that more data only improves forecasting accuracy up to a certain point. Thus, having more sensors appears to be more valuable than increasing the time resolution of measurements." ++++ diff --git a/content/publications/the-vision-of-bigbench-2-0.md b/content/publications/the-vision-of-bigbench-2-0.md new file mode 100644 index 0000000..6db128a --- /dev/null +++ b/content/publications/the-vision-of-bigbench-2-0.md @@ -0,0 +1,10 @@ ++++ +title = "The Vision of BigBench 2.0" +year = 2015 +authors = ["Tilmann Rabl", "Michael Frank", "Manuel Danisch", "Hans-Arno Jacobsen", "Bhaskar Gowda"] +venue = "Proceedings of the Fourth Workshop on Data analytics in the Cloud" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/2799562.2799642" +abstract = "Data is one of the most important resources for modern enterprises. Better analytics allow for a better understanding of customer requirements and market dynamics. The more data is collected, the more information can be extracted. However, information value extraction is limited by data processing speeds. Due to fast technological advances in big data management there is an abundance of big data systems. This leaves users in the dilemma of choosing a system that features good end-to-end performance for the use case. To get a good understanding of the actual performance of a system, realistic application level workloads are required." ++++ diff --git a/content/publications/total-order-in-content-based-publish-subscribe-systems.md b/content/publications/total-order-in-content-based-publish-subscribe-systems.md new file mode 100644 index 0000000..d1b03a4 --- /dev/null +++ b/content/publications/total-order-in-content-based-publish-subscribe-systems.md @@ -0,0 +1,10 @@ ++++ +title = "Total Order in Content-Based Publish/Subscribe Systems" +year = 2012 +authors = ["Kaiwen Zhang", "Vinod Muthusamy", "Hans-Arno Jacobsen"] +venue = "2012 IEEE 32nd International Conference on Distributed Computing Systems" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icdcs.2012.17" +abstract = "Total ordering is a messaging guarantee increasingly required of content-based pub/sub systems, which are traditionally focused on performance. The main challenge is the uniform ordering of streams of publications from multiple publishers within an overlay broker network to be delivered to multiple subscribers. Our solution integrates total ordering into the pub/sub logic instead of offloading it as an external service. We show that our solution is fully distributed and relies only on local broker knowledge and overlay links. We can identify and isolate specific publications and subscribers where synchronization is required: the overhead is therefore contained to the affected subscribers. Our solution remains safe under the presence of failure, where we show total order to be impossible to maintain. Our experiments demonstrate that our solution scales with the number of subscriptions and has limited overhead for the non-conflicting cases. A holistic comparison with group communication systems is offered to evaluate their relative scalability." ++++ diff --git a/content/publications/toward-intelligent-sustainable-and-reliable-cloud-database-systems-a-unified-research-vision.md b/content/publications/toward-intelligent-sustainable-and-reliable-cloud-database-systems-a-unified-research-vision.md new file mode 100644 index 0000000..c22fd84 --- /dev/null +++ b/content/publications/toward-intelligent-sustainable-and-reliable-cloud-database-systems-a-unified-research-vision.md @@ -0,0 +1,9 @@ ++++ +title = "Toward Intelligent, Sustainable, and Reliable Cloud Database Systems: A Unified Research Vision" +year = 2025 +authors = ["Michail Bachras", "Michael Dang'ana", "Yunhao Mao", "Shiquan Zhang", "Yuqiu Zhang", "Hans-Arno Jacobsen"] +venue = "2025 IEEE International Conference on Collaborative Advances in Software and COmputiNg (CASCON)" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/cascon66301.2025.00122" ++++ diff --git a/content/publications/towards-a-complete-bigbench-implementation.md b/content/publications/towards-a-complete-bigbench-implementation.md new file mode 100644 index 0000000..7d96df4 --- /dev/null +++ b/content/publications/towards-a-complete-bigbench-implementation.md @@ -0,0 +1,9 @@ ++++ +title = "Towards a Complete BigBench Implementation" +year = 2015 +authors = ["Tilmann Rabl", "Michael Frank", "Manuel Danisch", "Bhaskar Gowda", "Hans-Arno Jacobsen"] +venue = "Lecture Notes in Computer Science; Big Data Benchmarking" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1007/978-3-319-20233-4_1" ++++ diff --git a/content/publications/towards-an-optimally-distributed-quantum-fourier-transform-circuit.md b/content/publications/towards-an-optimally-distributed-quantum-fourier-transform-circuit.md new file mode 100644 index 0000000..f0df3c4 --- /dev/null +++ b/content/publications/towards-an-optimally-distributed-quantum-fourier-transform-circuit.md @@ -0,0 +1,10 @@ ++++ +title = "Towards an Optimally Distributed Quantum Fourier Transform Circuit" +year = 2026 +authors = ["Zachary Vernec", "Michael Silver", "Hans-Arno Jacobsen"] +venue = "arXiv" +publication_type = "ArXiv Preprint" +research = ["quantum-computing-systems"] +external_url = "https://arxiv.org/abs/2606.18494" +abstract = "A promising avenue for scaling quantum computing is to connect quantum processing units (QPUs) by generating entanglement between them. This requires circuit partitioning: partially rewriting quantum circuits to run on a distributed quantum system using quantum teleportation protocols, while preserving the unitary operation implemented by the circuit. The key metric to minimize when partitioning is the e-bit count, defined as the number of maximally entangled qubit pairs that must be generated between QPUs. We focus on partitioning the quantum Fourier transform (QFT) circuit, which is widely used as a subroutine in quantum algorithms such as quantum phase estimation and arithmetic circuits. Specifically, we present a partitioning scheme based on optimal gate-packing, compare it against prior analytical partitioning schemes for the QFT, and evaluate it against partitions produced by general-purpose circuit partitioning algorithms. We further validate our approach by implementing the partitioned circuit on quantum hardware." ++++ diff --git a/content/publications/towards-dependable-scalable-and-pervasive-distributed-ledgers-with-blockchains.md b/content/publications/towards-dependable-scalable-and-pervasive-distributed-ledgers-with-blockchains.md new file mode 100644 index 0000000..3325067 --- /dev/null +++ b/content/publications/towards-dependable-scalable-and-pervasive-distributed-ledgers-with-blockchains.md @@ -0,0 +1,10 @@ ++++ +title = "Towards Dependable, Scalable, and Pervasive Distributed Ledgers with Blockchains" +year = 2018 +authors = ["Kaiwen Zhang", "Hans-Arno Jacobsen"] +venue = "2018 IEEE 38th International Conference on Distributed Computing Systems (ICDCS)" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icdcs.2018.00134" +abstract = "Distributed blockchain ledgers are on the verge of becoming a disruptive technology, profoundly impacting a wide range of industries and established applications, such as cryptocurrency, and allowing for novel use cases in both the public sector (e.g., eGovernment, eHealth, etc.) and the private sector (e.g., finance, supply chain management, etc.). Blockchains promise the ability to maintain critical information in a trustworthy repository without any centralized management. The reliability of blockchain-enabled applications is based on the innate immutability of stored data, maintained through cryptographic means, which enables blockchains to provide transparency, efficiency, auditability, trust, and security. As the technology is still in its infancy, a number of pain points must be addressed in order to make distributed ledgers more dependable, scalable, and pervasive. In this paper, we present the research landscape in distributed ledger technology (DLT). To do so, we describe a taxonomy of blockchain applications called blockchain generations. We also present the DCS properties (Decentralization, Consistency, and Scalability) as an analogy to the CAP theorem. Furthermore, we provide a general structure of the blockchain platform which decomposes the distributed ledger into six layers: Application, Modeling, Contract, System, Data, and Network. Finally, we classify research angles across three dimensions: DCS properties impacted, targeted applications, and related layers." ++++ diff --git a/content/publications/towards-highly-parallel-event-processing-through-reconfigurable-hardware.md b/content/publications/towards-highly-parallel-event-processing-through-reconfigurable-hardware.md new file mode 100644 index 0000000..bbfbbb4 --- /dev/null +++ b/content/publications/towards-highly-parallel-event-processing-through-reconfigurable-hardware.md @@ -0,0 +1,10 @@ ++++ +title = "Towards highly parallel event processing through reconfigurable hardware" +year = 2011 +authors = ["Mohammad Sadoghi", "Harsh Singh", "Hans-Arno Jacobsen"] +venue = "Proceedings of the Seventh International Workshop on Data Management on New Hardware" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/1995441.1995445" +abstract = "We present fpga-ToPSS (Toronto Publish/Subscribe System), an efficient event processing platform to support high-frequency and low-latency event matching. fpga-ToPSS is built over reconfigurable hardware---FPGAs---to achieve line-rate processing by exploring various degrees of parallelism. Furthermore, each of our proposed FPGA-based designs is geared towards a unique application requirement, such as flexibility, adaptability, scalability, or pure performance, such that each solution is specifically optimized to attain a high level of parallelism. Therefore, each solution is formulated as a design trade-off between the degree of parallelism versus the desired application requirement. Moreover, our event processing engine supports Boolean expression matching with an expressive predicate language applicable to a wide range of applications including real-time data analysis, algorithmic trading, targeted advertisement, and (complex) event processing." ++++ diff --git a/content/publications/towards-just-in-time-middleware-architectures.md b/content/publications/towards-just-in-time-middleware-architectures.md new file mode 100644 index 0000000..85fa016 --- /dev/null +++ b/content/publications/towards-just-in-time-middleware-architectures.md @@ -0,0 +1,10 @@ ++++ +title = "Towards just-in-time middleware architectures" +year = 2005 +authors = ["Charles Zhang", "Dapeng Gao", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 4th international conference on Aspect-oriented software development" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/1052898.1052904" +abstract = "Middleware becomes increasingly important in building distributed applications. Today, conventional middleware systems are designed, implemented, and packaged prior to their applications. We argue that with this middleware construction paradigm it is often difficult to meet the challenges imposed by application specific customization requirements. We propose to reverse this paradigm by automatically synthesizing middleware structures as the result of reasoning about the distribution needs of the user application of middleware. We term this type of post-postulated middleware Just-in-time middleware (JiM). In this paper, we present our initial design and present an evaluation of the JiM paradigm through Abacus, a CORBA middleware implementation based on the aspect oriented refactoring of an industrial strength object request broker. In addition, we present Arachne, the Abacus synthesizer, which integrates source analysis, feature inference, and implementation synthesis. Our evaluations show that, through automatic synthesis alone, Abacus is able to support diversified application domains with very flexible architectural compositions and versatile resource requirements as compared to conventional pre-postulated approaches." ++++ diff --git a/content/publications/towards-planning-the-transformation-of-overlays.md b/content/publications/towards-planning-the-transformation-of-overlays.md new file mode 100644 index 0000000..ef9d739 --- /dev/null +++ b/content/publications/towards-planning-the-transformation-of-overlays.md @@ -0,0 +1,10 @@ ++++ +title = "Towards Planning the Transformation of Overlays" +year = 2015 +authors = ["Young Yoon", "Nathan Robinson", "Vinod Muthusamy", "Sheila A. McIlraith", "Hans-Arno Jacobsen"] +venue = "2015 IEEE 35th International Conference on Distributed Computing Systems" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icdcs.2015.107" +abstract = "Reconfiguring a topology is an important management technique to sustain high efficiency and robustness of an overlay. But, the problem of transforming the overlay from an old topology to a newly refined topology, at runtime, has received relatively little attention. The key challenge is to minimize the disruption that can be caused by topology transformation operations. Excessive disruption can be costly and harmful and thus it may hamper the decision to migrate to a better topology. To address this issue, we solve a problem of finding an appropriate sequence of steps to transform a topology that incurs the least service disruption. We refer to this problem as an incremental topology transformation (ITT) problem. The ITT problem can be formulated as an automated planning problem and can be solved with numerous off-the-shelf planning techniques. However, we found that state-of-the-art domain-independent planning techniques did not scale to solve large ITT problem instances. This shortcoming motivated us to develop a suite of planners that use novel domain-specific heuristics to guide the search for a solution. We empirically evaluated our planners on a wide range of topologies. Our results illustrate that our planners offer a viable solution to a diversity of ITT problems. We envision that our approach could eventually provide a compelling addition to the arsenal of techniques currently employed by the administrators of distributed overlay networks." ++++ diff --git a/content/publications/towards-scalable-publish-subscribe-systems.md b/content/publications/towards-scalable-publish-subscribe-systems.md new file mode 100644 index 0000000..828ea8e --- /dev/null +++ b/content/publications/towards-scalable-publish-subscribe-systems.md @@ -0,0 +1,10 @@ ++++ +title = "Towards Scalable Publish/Subscribe Systems" +year = 2015 +authors = ["Shuping Ji", "Chunyang Ye", "Jun Wei", "Hans-Arno Jacobsen"] +venue = "2015 IEEE 35th International Conference on Distributed Computing Systems" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icdcs.2015.108" +abstract = "Despite suffering from inefficiency and flexibility limitations, the filter-based routing (FBR) algorithm is widely used in content-based publish/subscribe (pub/sub) systems. To address its limitations, we propose a dynamic destination-based routing algorithm called D-DBR, which decomposes pub/sub into two independent parts: Content-based matching and destination based multicasting. D-DBR exhibits low event matching cost and high efficiency, flexibility, and robustness for event routing in small-scale overlays. To improve its scalability to large-scale overlays, we further extend D-DBR to a new routing algorithm called MERC. MERC divides the overlay into interconnected clusters and applies content-based and destination-based mechanisms to route events inter- and intra-cluster, respectively. We implemented all algorithms in the PADRES pub/sub system. Experimental results show that our algorithms outperform the FBR algorithm." ++++ diff --git a/content/publications/towards-solving-the-data-availability-problem-for-sharded-ethereum.md b/content/publications/towards-solving-the-data-availability-problem-for-sharded-ethereum.md new file mode 100644 index 0000000..5bf8739 --- /dev/null +++ b/content/publications/towards-solving-the-data-availability-problem-for-sharded-ethereum.md @@ -0,0 +1,10 @@ ++++ +title = "Towards Solving the Data Availability Problem for Sharded Ethereum" +year = 2018 +authors = ["Daniel Sel", "Kaiwen Zhang", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 2nd Workshop on Scalable and Resilient Infrastructures for Distributed Ledgers" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/3284764.3284769" +abstract = "The success and growing popularity of blockchain technology has lead to a significant increase in load on popular permissionless blockchains such as Ethereum. With the current design, these blockchain systems do not scale with additional nodes since every node executes every transaction. Further efforts are therefore necessary to develop scalable permissionless blockchain systems." ++++ diff --git a/content/publications/towards-vulnerability-based-intrusion-detection-with-event-processing.md b/content/publications/towards-vulnerability-based-intrusion-detection-with-event-processing.md new file mode 100644 index 0000000..664b588 --- /dev/null +++ b/content/publications/towards-vulnerability-based-intrusion-detection-with-event-processing.md @@ -0,0 +1,10 @@ ++++ +title = "Towards vulnerability-based intrusion detection with event processing" +year = 2011 +authors = ["Amer Farroukh", "Mohammad Sadoghi", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 5th ACM international conference on Distributed event-based system" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/2002259.2002284" +abstract = "Computer systems continue to be breached despite substantial investments in defense mechanisms to stop attacks from propagating. The accuracy of current intrusion detection systems (IDSes) is hindered by the limited capability of regular expressions (REs) to express the exact vulnerability. Recent advances have proposed vulnerability-based IDSes that parse traffic and retrieve protocol semantics to describe the vulnerability. Such a description of attacks is analogous to subscriptions that specify events of interest in event processing systems. However, the matching engine of state-of-the-art IDSes lacks efficient matching algorithms that can process many signatures simultaneously. In this work, we place event processing in the core of the IDS and propose novel algorithms to efficiently match vulnerability signatures. Also, we are among the first to detect complex attacks such as the Conficker worm which requires correlating multiple protocol data units (MPDUs) while maintaining a small memory footprint. Finally, we show that our algorithms are resilient to attacks through extensive testing of the IDS under different workloads. Our approach incurs negligible overhead when processing clean traffic and is faster than existing systems." ++++ diff --git a/content/publications/tpc-di-the-first-industry-benchmark-for-data-integration.md b/content/publications/tpc-di-the-first-industry-benchmark-for-data-integration.md new file mode 100644 index 0000000..86c75a9 --- /dev/null +++ b/content/publications/tpc-di-the-first-industry-benchmark-for-data-integration.md @@ -0,0 +1,10 @@ ++++ +title = "TPC-DI: The First Industry Benchmark for Data Integration" +year = 2014 +authors = ["Meikel Poess", "Tilmann Rabl", "Hans-Arno Jacobsen", "Brian Caufield"] +venue = "Proceedings of the VLDB Endowment" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.14778/2733004.2733009" +abstract = "Historically, the process of synchronizing a decision support system with data from operational systems has been referred to as Extract, Transform, Load (ETL) and the tools supporting such process have been referred to as ETL tools. Recently, ETL was replaced by the more comprehensive acronym, data integration (DI). DI describes the process of extracting and combining data from a variety of data source formats, transforming that data into a unified data model representation and loading it into a data store. This is done in the context of a variety of scenarios, such as data acquisition for business intelligence, analytics and data warehousing, but also synchronization of data between operational applications, data migrations and conversions, master data management, enterprise data sharing and delivery of data services in a service-oriented architecture context, amongst others. With these scenarios relying on up-to-date information it is critical to implement a highly performing, scalable and easy to maintain data integration system. This is especially important as the complexity, variety and volume of data is constantly increasing and performance of data integration systems is becoming very critical. Despite the significance of having a highly performing DI system, there has been no industry standard for measuring and comparing their performance. The TPC, acknowledging this void, has released TPC-DI, an innovative benchmark for data integration. This paper motivates the reasons behind its development, describes its main characteristics including workload, run rules, metric, and explains key decisions." ++++ diff --git a/content/publications/transactional-mobility-in-distributed-content-based-publish-subscribe-systems.md b/content/publications/transactional-mobility-in-distributed-content-based-publish-subscribe-systems.md new file mode 100644 index 0000000..2232819 --- /dev/null +++ b/content/publications/transactional-mobility-in-distributed-content-based-publish-subscribe-systems.md @@ -0,0 +1,10 @@ ++++ +title = "Transactional Mobility in Distributed Content-Based Publish/Subscribe Systems" +year = 2009 +authors = ["Songlin Hu", "Vinod Muthusamy", "Guoli Li", "Hans-Arno Jacobsen"] +venue = "2009 29th IEEE International Conference on Distributed Computing Systems" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icdcs.2009.73" +abstract = "This paper formalizes transactional properties for publish/subscribe client mobility and develops protocols to realize them. Evaluations show that compared to traditional protocols, those developed in this paper, in addition to supporting transactional properties, are more stable with respect to message and processing overheads. Changes in factors such as the number of moving clients have little impact, making the protocols more scalable and simpler to administer due to predictable resource requirements." ++++ diff --git a/content/publications/transfer-learning-with-time-series-data-a-systematic-mapping-study.md b/content/publications/transfer-learning-with-time-series-data-a-systematic-mapping-study.md new file mode 100644 index 0000000..54d3bf9 --- /dev/null +++ b/content/publications/transfer-learning-with-time-series-data-a-systematic-mapping-study.md @@ -0,0 +1,10 @@ ++++ +title = "Transfer Learning With Time Series Data: A Systematic Mapping Study" +year = 2021 +authors = ["Manuel Weber", "Maximilian Auch", "Christoph Doblander", "Peter Mandl", "Hans-Arno Jacobsen"] +venue = "IEEE Access" +publication_type = "Journal Article" +research = ["distributed-machine-learning"] +external_url = "https://doi.org/10.1109/access.2021.3134628" +abstract = "Transfer Learning is a well-studied concept in machine learning, that relaxes the assumption that training and testing data need to be drawn from the same distribution. Recent success in applying transfer learning in the area of computer vision has motivated research on transfer learning also in context of time series data. This benefits learning in various time series domains, including a variety of domains based on sensor values. In this paper, we conduct a systematic mapping study of literature on transfer learning with time series data. Following the review guidelines of Kitchenham and Charters, we identify and analyze 223 relevant publications. We describe the pursued approaches and point out trends. Especially during the last two years, there has been a vast increase in the number of publications on the topic. This paper’s findings can help researchers as well as practitioners getting into the field and can help identify research gaps." ++++ diff --git a/content/publications/two-perspectives-on-blockchains-capabilities-vs-features.md b/content/publications/two-perspectives-on-blockchains-capabilities-vs-features.md new file mode 100644 index 0000000..e37683b --- /dev/null +++ b/content/publications/two-perspectives-on-blockchains-capabilities-vs-features.md @@ -0,0 +1,10 @@ ++++ +title = "Two Perspectives on Blockchains: Capabilities vs. Features" +year = 2019 +authors = ["Søren Debois", "Marlon Dumas", "Stephan Haarmann", "Hans-Arno Jacobsen", "Mieke Jans", "Jan Mendling", "Mark Staples", "Barbara Weber", "Francesca Zerbato", "Kaiwen Zhang"] +venue = "Blockchain Technology for Collaborative Information Systems (Dagstuhl Seminar 18332), Dagstuhl Reports 8, pp. 82–88" +publication_type = "Technical Report" +research = ["data-management"] +external_url = "https://documentserver.uhasselt.be/handle/1942/28120" +abstract = "Blockchain technology enables an evolving set of parties to maintain a safe, permanent, and tamper-proof ledger of transactions without a central authority. This technology opens manifold opportunities to redesign business-to-business collaborations, while bringing about numerous challenges. These opportunities and challenges were discussed in the Dagstuhl Seminar 18332 “Blockchain Technology for Collaborative Information Systems”. This report documents the program and the outcomes of the seminar." ++++ diff --git a/content/publications/using-publish-subscribe-middleware-for-distributed-ev-charging-optimization.md b/content/publications/using-publish-subscribe-middleware-for-distributed-ev-charging-optimization.md new file mode 100644 index 0000000..d7bd64d --- /dev/null +++ b/content/publications/using-publish-subscribe-middleware-for-distributed-ev-charging-optimization.md @@ -0,0 +1,9 @@ ++++ +title = "Using publish/subscribe middleware for distributed EV charging optimization" +year = 2016 +authors = ["José Rivera", "Martin Jergler", "Aleksandar Stoimenov", "Christoph Goebel", "Hans-Arno Jacobsen"] +venue = "Computer Science - Research and Development" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.1007/s00450-014-0278-4" ++++ diff --git a/content/publications/v-guard.md b/content/publications/v-guard.md index 940dc26..32a8632 100644 --- a/content/publications/v-guard.md +++ b/content/publications/v-guard.md @@ -1,16 +1,11 @@ +++ -title = "V-Guard: An Efficient Permissioned Blockchain for Achieving Consensus under Dynamic Memberships in V2X Networks" +title = "V-Guard: An Efficient Permissioned Blockchain for Achieving Consensus under Dynamic Memberships in V2X" year = 2023 authors = ["Gengrui Zhang", "Yunhao Mao", "Shiquan Zhang", "Shashank Motepalli", "Fei Pan", "Hans-Arno Jacobsen"] -venue = "arXiv Preprint" +venue = "arXiv" publication_type = "ArXiv Preprint" research = ["data-management"] tags = ["blockchain", "consensus", "v2x"] -summary = "Permissioned blockchain design for vehicular networks with changing memberships and intermittent connectivity." external_url = "https://arxiv.org/abs/2301.06210" +abstract = "This paper presents V-Guard, a new permissioned blockchain that achieves consensus for vehicular data under changing memberships, targeting the problem in V2X networks where vehicles are often intermittently connected on the roads. To achieve this goal, V-Guard integrates membership management into the consensus process for agreeing on data entries. It binds a data entry with a membership configuration profile that describes responsible vehicles for achieving consensus for the data entry. As such, V-Guard produces chained consensus results of both data entries and their residing membership profiles, which enables consensus to be achieved seamlessly under changing memberships. In addition, V-Guard separates the ordering of transactions from consensus, allowing concurrent ordering instances and periodic consensus instances to order and commit data entries. These features make V-Guard efficient for achieving consensus under dynamic memberships with high throughput and latency performance." +++ - -V-Guard targets consensus in V2X settings where participation changes -constantly. The design folds membership management into consensus so the system -can keep ordering and committing data even as the responsible set of vehicles -changes over time. diff --git a/content/publications/variations-of-the-star-schema-benchmark-to-test-the-effects-of-data-skew-on-query-performance.md b/content/publications/variations-of-the-star-schema-benchmark-to-test-the-effects-of-data-skew-on-query-performance.md new file mode 100644 index 0000000..b0f511f --- /dev/null +++ b/content/publications/variations-of-the-star-schema-benchmark-to-test-the-effects-of-data-skew-on-query-performance.md @@ -0,0 +1,10 @@ ++++ +title = "Variations of the star schema benchmark to test the effects of data skew on query performance" +year = 2013 +authors = ["Tilmann Rabl", "Meikel Poess", "Hans-Arno Jacobsen", "Patrick E. O'Neil", "Elizabeth J. O'Neil"] +venue = "Proceedings of the 4th ACM/SPEC International Conference on Performance Engineering" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/2479871.2479927" +abstract = "The Star Schema Benchmark (SSB), now in its third revision, has been widely used to evaluate the performance of database management systems when executing star schema queries. SSB, based on the well known industry standard benchmark TPC-H, shares some of its drawbacks, most notably, its uniform data distributions. Today's systems rely heavily on sophisticated cost-based query optimizers to generate the most efficient query execution plans. A benchmark that evaluates optimizer's capability to generate optimal execution plans under all circumstances must provide the rich data set details on which optimizers rely (uniform and non-uniform distributions, data sparsity, etc.). This is also true for other database system parts, such as indices and operators, and ultimately holds for an end-to-end benchmark as well. SSB's data generator, based on TPC-H's dbgen, is not easy to adapt to different data distributions as its meta data and actual data generation implementations are not separated. In this paper, we motivate the need for a new revision of SSB that includes non-uniform data distributions. We list what specific modifications are required to SSB to implement non-uniform data sets and we demonstrate how to implement these modifications in the Parallel Data Generator Framework to generate both the data and query sets." ++++ diff --git a/content/publications/vehicle-originating-signals-for-real-time-charging-control-of-electric-vehicle-fleets.md b/content/publications/vehicle-originating-signals-for-real-time-charging-control-of-electric-vehicle-fleets.md new file mode 100644 index 0000000..c381500 --- /dev/null +++ b/content/publications/vehicle-originating-signals-for-real-time-charging-control-of-electric-vehicle-fleets.md @@ -0,0 +1,10 @@ ++++ +title = "Vehicle-Originating-Signals for Real-Time Charging Control of Electric Vehicle Fleets" +year = 2015 +authors = ["Victor del Razo", "Christoph Goebel", "Hans-Arno Jacobsen"] +venue = "IEEE Transactions on Transportation Electrification" +publication_type = "Journal Article" +research = [] +external_url = "https://doi.org/10.1109/tte.2015.2445215" +abstract = "We propose the vehicle-originating-signals (VOS) approach for charging control of a fleet of electric vehicles (EVs) in an electricity distribution network (DN). The goal of the approach is to manage the EV load such that the aggregate power consumption, including inflexible demand and solar generation, closely follows a given target profile. The VOS approach enables EVs in a fleet to compute signals reflecting their need for charge and willingness to supply power. An aggregator collects these signals and implements the control with minor computational effort. We evaluate the VOS approach on a scenario for Munich, Germany, based on a mobility survey and real electricity demand and solar generation data. First, we compare our approach against a centralized optimization in terms of objective fulfillment and solving time. We show that it achieves a competitive performance, especially for large vehicle fleets. Second, we present a statistical method to evaluate the performance and limitations of our approach. We identify a recommended range of the vehicle fleet size for a given load magnitude and show that the target profiles can be met with a defined certainty. Finally, we present a method for further reducing the communication overhead with minor effect on performance." ++++ diff --git a/content/publications/waveform-signal-entropy-and-compression-study-of-whole-building-energy-datasets.md b/content/publications/waveform-signal-entropy-and-compression-study-of-whole-building-energy-datasets.md new file mode 100644 index 0000000..88b76a3 --- /dev/null +++ b/content/publications/waveform-signal-entropy-and-compression-study-of-whole-building-energy-datasets.md @@ -0,0 +1,10 @@ ++++ +title = "Waveform Signal Entropy and Compression Study of Whole-Building Energy Datasets" +year = 2019 +authors = ["Thomas Kriechbaumer", "Daniel Jorde", "Hans-Arno Jacobsen"] +venue = "Proceedings of the Tenth ACM International Conference on Future Energy Systems" +publication_type = "Conference Paper" +research = [] +external_url = "https://doi.org/10.1145/3307772.3328285" +abstract = "Electrical energy consumption has been an ongoing research area since the coming of smart homes and Internet of Things devices. Consumption characteristics and usages profiles are directly influenced by building occupants and their interaction with electrical appliances. Extracted information from these data can be used to conserve energy and increase user comfort levels. Data analysis together with machine learning models can be utilized to extract valuable information for the benefit of occupants themselves, power plants, and grid operators. Public energy datasets provide a scientific foundation to develop and benchmark these algorithms and techniques. With datasets exceeding tens of terabytes, we present a novel study of five whole-building energy datasets with high sampling rates, their signal entropy, and how a well-calibrated measurement can have a significant effect on the overall storage requirements. We show that some datasets do not fully utilize the available measurement precision, therefore leaving potential accuracy and space savings untapped. We benchmark a comprehensive list of 365 file formats, transparent data transformations, and lossless compression algorithms. The primary goal is to reduce the overall dataset size while maintaining an easy-to-use file format and access API. We show that with careful selection of file format and encoding scheme, we can reduce the size of some datasets by up to 73%." ++++ diff --git a/content/publications/wavegas-waveform-relaxation-for-scaling-graph-neural-networks.md b/content/publications/wavegas-waveform-relaxation-for-scaling-graph-neural-networks.md new file mode 100644 index 0000000..5495f9a --- /dev/null +++ b/content/publications/wavegas-waveform-relaxation-for-scaling-graph-neural-networks.md @@ -0,0 +1,10 @@ ++++ +title = "WaveGAS: Waveform Relaxation for Scaling Graph Neural Networks" +year = 2025 +authors = ["Jana Vatter", "Mykhaylo Zayats", "Marcos Martínez Galindo", "Vanessa López", "Ruben Mayer", "Hans-Arno Jacobsen", "Hoang Thanh Lam"] +venue = "arXiv" +publication_type = "ArXiv Preprint" +research = ["distributed-machine-learning"] +external_url = "https://arxiv.org/abs/2502.19986" +abstract = "With the ever-growing size of real-world graphs, numerous techniques to overcome resource limitations when training Graph Neural Networks (GNNs) have been developed. One such approach, GNNAutoScale (GAS), uses graph partitioning to enable training under constrained GPU memory. GAS also stores historical embedding vectors, which are retrieved from one-hop neighbors in other partitions, ensuring critical information is captured across partition boundaries. The historical embeddings which come from the previous training iteration are stale compared to the GAS estimated embeddings, resulting in approximation errors of the training algorithm. Furthermore, these errors accumulate over multiple layers, leading to suboptimal node embeddings. To address this shortcoming, we propose two enhancements: first, WaveGAS, inspired by waveform relaxation, performs multiple forward passes within GAS before the backward pass, refining the approximation of historical embeddings and gradients to improve accuracy; second, a gradient-tracking method that stores and utilizes more accurate historical gradients during training. Empirical results show that WaveGAS enhances GAS and achieves better accuracy, even outperforming methods that train on full graphs, thanks to its robust estimation of node embeddings." ++++ diff --git a/content/publications/weighted-overlay-design-for-topic-based-publish-subscribe-systems-on-geo-distributed-data-centers.md b/content/publications/weighted-overlay-design-for-topic-based-publish-subscribe-systems-on-geo-distributed-data-centers.md new file mode 100644 index 0000000..fa277cb --- /dev/null +++ b/content/publications/weighted-overlay-design-for-topic-based-publish-subscribe-systems-on-geo-distributed-data-centers.md @@ -0,0 +1,10 @@ ++++ +title = "Weighted Overlay Design for Topic-Based Publish/Subscribe Systems on Geo-Distributed Data Centers" +year = 2015 +authors = ["Chen Chen", "Yoav Tock", "Hans-Arno Jacobsen", "Roman Vitenberg"] +venue = "2015 IEEE 35th International Conference on Distributed Computing Systems" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icdcs.2015.55" +abstract = "We incorporate underlay information into overlay design for topic-based publish/subscribe (pub/sub) systems on geo-distributed data centers. We propose the MinAvg-WTCO problem that optimizes the weighted average node degree while constructing a topic-connected overlay (TCO), i.e., Each topic induces a connected sub-overlay among all nodes interested in this topic. Most existing TCO designs are oblivious to the low-level network infrastructure and assume edge equivalence. We prove that MinAvg-WTCO is NP-complete and difficult to approximate within a logarithmic factor with regard to the number of nodes. We devise several approximation algorithms for MinAvg-WTCO using different design techniques. Both theoretical analysis and empirical evaluation show that our designed algorithms tread the balance between overlay quality and runtime cost. Our algorithms significantly outperform the state of the art for TCO design that ignores edge differences." ++++ diff --git a/content/publications/when-agentic-executions-fail-detecting-and-localizing-runtime-faults-from-telemetry.md b/content/publications/when-agentic-executions-fail-detecting-and-localizing-runtime-faults-from-telemetry.md new file mode 100644 index 0000000..99e1ca3 --- /dev/null +++ b/content/publications/when-agentic-executions-fail-detecting-and-localizing-runtime-faults-from-telemetry.md @@ -0,0 +1,10 @@ ++++ +title = "When Agentic Executions Fail: Detecting and Localizing Runtime Faults from Telemetry" +year = 2026 +authors = ["Chenkai Zhang", "Yiran Li", "Yifang Tian", "Michalis Bachras", "Hans-Arno Jacobsen"] +venue = "arXiv" +publication_type = "ArXiv Preprint" +research = ["distributed-machine-learning"] +external_url = "https://arxiv.org/abs/2608.14680" +abstract = "Reliability in LLM-based agentic systems is a property of the whole execution (its tool calls, model calls, guardrails, and inter-agent messages), not of the final answer alone, yet evaluating only task outcomes reveals little about how or why a run fails. We present AGENTCHAOSBENCH, a benchmark for detecting and localizing runtime faults in agentic systems from their execution telemetry. We run five heterogeneous applications that coordinate agents over the Agent-to-Agent protocol and call tools through the Model Context Protocol, and inject ten types of operational fault (unavailable or slow tools, corrupted or oversized responses, and delayed, looped, or misrouted delegations and bypassed guardrails) at their tool, model, guardrail, and inter-agent boundaries, alongside a no-fault control. The resulting dataset contains 275 sanitized traces: 250 faulty executions spanning ten fault types and 25 no-fault controls. Each faulty trace is aligned with the no-fault execution of the same input; fault-type labels and, where applicable, location labels are held out from diagnosis. On structured single-trace inputs, a first set of zero-shot LLM baselines shows the task is far from solved: local detectors up to 14B parameters reach only 13.6-19.2% top-1 fault-type accuracy and the frontier DeepSeek-v4-pro only 24.8%, while jointly identifying the fault type and its location tops out at 22%; reference-dependent faults (above all a bypassed guardrail) stay near-unsolved from a single trace. An aligned reference improves selected relative faults but does not resolve guardrail bypass. The held-out labels and compact prediction format support reproducible comparison of LLM-based and non-LLM diagnosis methods." ++++ diff --git a/content/publications/where-is-my-training-bottleneck-hidden-trade-offs-in-deep-learning-preprocessing-pipelines.md b/content/publications/where-is-my-training-bottleneck-hidden-trade-offs-in-deep-learning-preprocessing-pipelines.md new file mode 100644 index 0000000..3abeae3 --- /dev/null +++ b/content/publications/where-is-my-training-bottleneck-hidden-trade-offs-in-deep-learning-preprocessing-pipelines.md @@ -0,0 +1,10 @@ ++++ +title = "Where Is My Training Bottleneck? Hidden Trade-Offs in Deep Learning Preprocessing Pipelines" +year = 2022 +authors = ["Alexander Isenko", "Ruben Mayer", "Jeffrey Jedele", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 2022 International Conference on Management of Data" +publication_type = "Conference Paper" +research = ["distributed-machine-learning"] +external_url = "https://doi.org/10.1145/3514221.3517848" +abstract = "Preprocessing pipelines in deep learning aim to provide sufficient data throughput to keep the training processes busy. Maximizing resource utilization is becoming more challenging as the throughput of training processes increases with hardware innovations (e.g., faster GPUs, TPUs, and inter-connects) and advanced parallelization techniques that yield better scalability. At the same time, the amount of training data needed in order to train increasingly complex models is growing. As a consequence of this development, data preprocessing and provisioning are becoming a severe bottleneck in end-to-end deep learning pipelines." ++++ diff --git a/content/publications/whitening-soa-testing-via-event-exposure.md b/content/publications/whitening-soa-testing-via-event-exposure.md new file mode 100644 index 0000000..2c612ec --- /dev/null +++ b/content/publications/whitening-soa-testing-via-event-exposure.md @@ -0,0 +1,10 @@ ++++ +title = "Whitening SOA Testing via Event Exposure" +year = 2013 +authors = ["Chunyang Ye", "Hans-Arno Jacobsen"] +venue = "IEEE Transactions on Software Engineering" +publication_type = "Journal Article" +research = ["data-management"] +external_url = "https://doi.org/10.1109/tse.2013.20" +abstract = "Whitening the testing of service-oriented applications can provide service consumers confidence on how well an application has been tested. However, to protect business interests of service providers and to prevent information leakage, the implementation details of services are usually invisible to service consumers. This makes it challenging to determine the test coverage of a service composition as a whole and design test cases effectively. To address this problem, we propose an approach to whiten the testing of service compositions based on events exposed by services. By deriving event interfaces to explore only necessary test coverage information from service implementations, our approach allows service consumers to determine test coverage based on selected events exposed by services at runtime without releasing the service implementation details. We also develop an approach to design test cases effectively based on event interfaces concerning both effectiveness and information leakage. The experimental results show that our approach outperforms existing testing approaches for service compositions with up to 49 percent more test coverage and an up to 24 percent higher fault-detection rate. Moreover, our solution can trade off effectiveness, efficiency, and information leakage for test case generation." ++++ diff --git a/content/publications/why-do-my-blockchain-transactions-fail-a-study-of-hyperledger-fabric.md b/content/publications/why-do-my-blockchain-transactions-fail-a-study-of-hyperledger-fabric.md new file mode 100644 index 0000000..eb8c1c6 --- /dev/null +++ b/content/publications/why-do-my-blockchain-transactions-fail-a-study-of-hyperledger-fabric.md @@ -0,0 +1,10 @@ ++++ +title = "Why Do My Blockchain Transactions Fail?: A Study of Hyperledger Fabric" +year = 2021 +authors = ["Jeeta Ann Chacko", "Ruben Mayer", "Hans-Arno Jacobsen"] +venue = "Proceedings of the 2021 International Conference on Management of Data" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1145/3448016.3452823" +abstract = "Permissioned blockchain systems promise to provide both decentralized trust and privacy. Hyperledger Fabric is currently one of the most wide-spread permissioned blockchain systems and is heavily promoted both in industry and academia. Due to its optimistic concurrency model, the transaction failure rates in Fabric can become a bottleneck. While there is active research to reduce failures, there is a lack of understanding on their root cause and, consequently, a lack of guidelines on how to configure Fabric optimally for different scenarios. To close this gap, in this paper, we first introduce a formal definition of the different types of transaction failures in Fabric. Then, we develop a comprehensive testbed and benchmarking system, HyperLedgerLab, along with four different chaincodes that represent realistic use cases and a chaincode/workload generator. Using HyperLedgerLab, we conduct exhaustive experiments to analyze the impact of different parameters of Fabric such as block size, endorsement policies, and others, on transaction failures. We further analyze three recently proposed optimizations from the literature, Fabric++, Streamchain and FabricSharp, and evaluate under which conditions they reduce the failure rates. Finally, based on our results, we provide recommendations for Fabric practitioners on how to configure the system and also propose new research directions." ++++ diff --git a/content/publications/xml-routing-in-data-dissemination-networks.md b/content/publications/xml-routing-in-data-dissemination-networks.md new file mode 100644 index 0000000..d3cf10d --- /dev/null +++ b/content/publications/xml-routing-in-data-dissemination-networks.md @@ -0,0 +1,10 @@ ++++ +title = "XML Routing in Data Dissemination Networks" +year = 2007 +authors = ["Guoli Li", "Shuang Hou", "Hans-Arno Jacobsen"] +venue = "2007 IEEE 23rd International Conference on Data Engineering" +publication_type = "Conference Paper" +research = ["data-management"] +external_url = "https://doi.org/10.1109/icde.2007.369021" +abstract = "This paper addresses the XML/XPath content-based routing problem. More specifically, this paper focuses on the problem of efficiently routing an XML document emitted from a data producers at one point in the network to a set of data consumers located anywhere throughout the network. Prior to receiving XML documents, consumers must have expressed interest in receiving XML documents by registering XPEs with the network. This problem statement is akin to the well-known publish/subscribe matching problem. However, the main difference here is that in the case of data dissemination networks there exists no one single centralized publish/subscribe system, but a network of content-based routers (i.e., a network or federation of publish/subscribe systems)." ++++ diff --git a/data/content_schema.json b/data/content_schema.json index 28a48cd..ff9bc01 100644 --- a/data/content_schema.json +++ b/data/content_schema.json @@ -347,13 +347,6 @@ "required": true, "format": "url" }, - { - "key": "source_url", - "label": "Metadata source", - "type": "string", - "default": "", - "format": "url" - }, { "key": "publication_type", "label": "Publication type", @@ -368,13 +361,6 @@ "default": [], "options_source": "research" }, - { - "key": "summary", - "label": "Short listing summary", - "type": "string", - "default": "", - "help": "Optional short description for the publication list; put the full abstract below." - }, { "key": "abstract", "label": "Abstract", @@ -384,6 +370,14 @@ "format": "markdown", "help": "Paste the paper's abstract. Use Markdown for emphasis and LaTeX for formulas: $x^2$ or \\(x^2\\) inline; $$...$$ or \\[...\\] for a separate equation. The formatting buttons help; save and open the local preview to check the result. Use code formatting for literal dollar amounts. Full LaTeX documents and custom packages are not supported." }, + { + "key": "abstract_license_url", + "label": "Abstract license", + "type": "string", + "default": "", + "format": "url", + "editor": false + }, { "key": "tags", "label": "Tags", diff --git a/docs/outstanding.md b/docs/outstanding.md index 90f1e48..26ad274 100644 --- a/docs/outstanding.md +++ b/docs/outstanding.md @@ -90,10 +90,11 @@ identicons, or other people's photos. (legacy spelling), `p2ptopss_workload_2009.tgz`, and `DatasetOfResourceDiscovery.zip`. Keep the site's unavailable-archive notices and contact links until the files are recovered. -- Obtain an authoritative group BibTeX/database export to check bibliography - completeness. The current archive contains the public legacy list and a - selection of newer papers. Preserve existing URLs when replacing preprints - with published versions. +- Obtain an authoritative group BibTeX/database export to check unindexed + bibliography gaps. The 2026-09-10 backfill added 336 papers and preferred + verified published versions; see `docs/publication-import.md` for sources, + version decisions, and the unresolved RSS filtering paper. Existing page + URLs were preserved. ## Website verification and administration diff --git a/docs/publication-abstracts-pending.json b/docs/publication-abstracts-pending.json new file mode 100644 index 0000000..0705806 --- /dev/null +++ b/docs/publication-abstracts-pending.json @@ -0,0 +1,192 @@ +[ + { + "paper": "a-bigbench-implementation-in-the-hadoop-ecosystem", + "title": "A BigBench Implementation in the Hadoop Ecosystem", + "source_url": "https://doi.org/10.1007/978-3-319-10596-3_1" + }, + { + "paper": "a-comprehensive-study-on-benchmarking-permissioned-blockchains", + "title": "A Comprehensive Study on Benchmarking Permissioned Blockchains", + "source_url": "https://doi.org/10.1007/978-3-031-68031-1_2" + }, + { + "paper": "a-generalized-algorithm-for-publish-subscribe-overlay-design-and-its-fast-implementation", + "title": "A Generalized Algorithm for Publish/Subscribe Overlay Design and Its Fast Implementation", + "source_url": "https://doi.org/10.1007/978-3-642-33651-5_6" + }, + { + "paper": "a-topss-a-publish-subscribe-system-supporting-approximate-matching", + "title": "A-TOPSS - A Publish/Subscribe System Supporting Approximate Matching", + "source_url": "https://doi.org/10.1016/b978-155860869-6/50120-7" + }, + { + "paper": "a-topss-a-publish-subscribe-system-supporting-imperfect-information-processing", + "title": "A-ToPSS: A Publish/Subscribe System Supporting Imperfect Information Processing", + "source_url": "https://doi.org/10.1016/b978-012088469-8.50127-3" + }, + { + "paper": "adversarial-robustness-in-distributed-quantum-machine-learning", + "title": "Adversarial Robustness in Distributed Quantum Machine Learning", + "source_url": "https://doi.org/10.1007/978-3-032-11153-1_11" + }, + { + "paper": "adversarial-robustness-of-partitioned-quantum-classifiers", + "title": "Adversarial Robustness of Partitioned Quantum Classifiers", + "source_url": "https://arxiv.org/abs/2502.20403" + }, + { + "paper": "big-data-generation", + "title": "Big Data Generation", + "source_url": "https://doi.org/10.1007/978-3-642-53974-9_3" + }, + { + "paper": "bigbench-specification-v0-1-bigbench-an-industry-standard-benchmark-for-big-data-analytics", + "title": "BigBench Specification V0.1 - BigBench: An Industry Standard Benchmark for Big Data Analytics", + "source_url": "https://doi.org/10.1007/978-3-642-53974-9_14" + }, + { + "paper": "bpm-in-cloud-architectures-business-process-management-with-slas-and-events", + "title": "BPM in Cloud Architectures: Business Process Management with SLAs and Events", + "source_url": "https://doi.org/10.1007/978-3-642-15618-2_2" + }, + { + "paper": "composite-subscriptions-in-content-based-publish-subscribe-systems", + "title": "Composite Subscriptions in Content-Based Publish/Subscribe Systems", + "source_url": "https://doi.org/10.1007/11587552_13" + }, + { + "paper": "discussion-of-bigbench-a-proposed-industry-standard-performance-benchmark-for-big-data", + "title": "Discussion of BigBench: A Proposed Industry Standard Performance Benchmark for Big Data", + "source_url": "https://doi.org/10.1007/978-3-319-15350-6_4" + }, + { + "paper": "efficient-and-scalable-filtering-of-graph-based-metadata", + "title": "Efficient and scalable filtering of graph-based metadata", + "source_url": "https://doi.org/10.1016/j.websem.2005.09.006" + }, + { + "paper": "efficient-constraint-processing-for-highly-personalized-location-based-services", + "title": "Efficient Constraint Processing for Highly Personalized Location Based Services", + "source_url": "https://doi.org/10.1016/b978-012088469-8.50128-5" + }, + { + "paper": "efficient-data-transfer-in-shared-storage-cloud-data-processing-systems-with-optics", + "title": "Efficient Data Transfer in Shared-storage Cloud Data Processing Systems with OPTICS", + "source_url": "https://dl.acm.org/doi/10.5555/3615924.3623630" + }, + { + "paper": "energieinformatik-aktuelle-und-zukunftige-forschungsschwerpunkte", + "title": "Energieinformatik - Aktuelle und zukünftige Forschungsschwerpunkte", + "source_url": "https://doi.org/10.1007/s11576-013-0396-9" + }, + { + "paper": "energy-informatics-current-and-future-research-directions", + "title": "Energy Informatics - Current and Future Research Directions", + "source_url": "https://doi.org/10.1007/s12599-013-0304-2" + }, + { + "paper": "eqosystem-supporting-fluid-distributed-service-oriented-workflows", + "title": "eQoSystem: supporting fluid distributed service-oriented workflows", + "source_url": "https://doi.org/10.1145/2002259.2002320" + }, + { + "paper": "event-exposure-for-web-services-a-grey-box-approach-to-compose-and-evolve-web-services", + "title": "Event Exposure for Web Services: A Grey-Box Approach to Compose and Evolve Web Services", + "source_url": "https://doi.org/10.1007/978-3-642-16599-3_14" + }, + { + "paper": "externalizing-java-server-concurrency-with-cal", + "title": "Externalizing Java Server Concurrency with CAL", + "source_url": "https://doi.org/10.1007/978-3-540-70592-5_16" + }, + { + "paper": "generic-middleware-substrate-through-modelware", + "title": "Generic Middleware Substrate Through Modelware", + "source_url": "https://doi.org/10.1007/11587552_16" + }, + { + "paper": "green-middleware", + "title": "Green Middleware", + "source_url": "https://doi.org/10.1007/978-3-642-22179-8_18" + }, + { + "paper": "nofare-a-non-intrusive-facility-resource-monitoring-system", + "title": "NoFaRe: A Non-Intrusive Facility Resource Monitoring System", + "source_url": "https://doi.org/10.1007/978-3-319-25876-8_6" + }, + { + "paper": "on-the-effects-of-signal-design-in-electric-vehicle-charging-using-vehicle-originating-signals", + "title": "On the effects of signal design in electric vehicle charging using vehicle-originating-signals", + "source_url": "https://doi.org/10.1007/s00450-014-0286-4" + }, + { + "paper": "opengridmap-an-open-platform-for-inferring-power-grids-with-crowdsourced-data", + "title": "OpenGridMap: An Open Platform for Inferring Power Grids with Crowdsourced Data", + "source_url": "https://doi.org/10.1007/978-3-319-25876-8_15" + }, + { + "paper": "opengridmap-towards-automatic-power-grid-simulation-model-generation-from-crowdsourced-data", + "title": "OpenGridMap: towards automatic power grid simulation model generation from crowdsourced data", + "source_url": "https://doi.org/10.1007/s00450-016-0317-4" + }, + { + "paper": "optimized-cluster-based-filtering-algorithm-for-graph-metadata", + "title": "Optimized cluster-based filtering algorithm for graph metadata", + "source_url": "https://doi.org/10.1016/j.ins.2011.08.002" + }, + { + "paper": "orchestrating-soa-using-requirement-specifications-and-domain-ontologies", + "title": "Orchestrating SOA Using Requirement Specifications and Domain Ontologies", + "source_url": "https://doi.org/10.1007/978-3-662-45391-9_30" + }, + { + "paper": "prism-is-research-in-aspect-mining", + "title": "PRISM is research in aSpect mining", + "source_url": "https://doi.org/10.1145/1028664.1028676" + }, + { + "paper": "processing-big-events-with-showers-and-streams", + "title": "Processing Big Events with Showers and Streams", + "source_url": "https://doi.org/10.1007/978-3-642-53974-9_6" + }, + { + "paper": "re-factoring-middleware-systems-a-case-study", + "title": "Re-factoring Middleware Systems: A Case Study", + "source_url": "https://doi.org/10.1007/978-3-540-39964-3_79" + }, + { + "paper": "remon-remote-external-memory-over-the-network", + "title": "REMON: Remote External Memory Over the Network", + "source_url": "https://doi.org/10.1109/icde65706.2026.00175" + }, + { + "paper": "service-subscription-and-consumption-for-personal-web-applications", + "title": "Service Subscription and Consumption for Personal Web Applications", + "source_url": "https://doi.org/10.1007/978-3-642-39995-4_3" + }, + { + "paper": "the-debs-2025-grand-challenge-real-time-monitoring-of-defects-in-laser-powder-bed-fusion-l-pbf-manufacturing", + "title": "The DEBS 2025 Grand Challenge: Real-Time Monitoring of Defects in Laser Powder Bed Fusion (L-PBF) Manufacturing", + "source_url": "https://doi.org/10.1145/3701717.3735578" + }, + { + "paper": "the-padres-distributed-publish-subscribe-system", + "title": "The PADRES Distributed Publish/Subscribe System", + "source_url": "https://dblp.org/rec/conf/fiw/FidlerJLM05" + }, + { + "paper": "toward-intelligent-sustainable-and-reliable-cloud-database-systems-a-unified-research-vision", + "title": "Toward Intelligent, Sustainable, and Reliable Cloud Database Systems: A Unified Research Vision", + "source_url": "https://doi.org/10.1109/cascon66301.2025.00122" + }, + { + "paper": "towards-a-complete-bigbench-implementation", + "title": "Towards a Complete BigBench Implementation", + "source_url": "https://doi.org/10.1007/978-3-319-20233-4_1" + }, + { + "paper": "using-publish-subscribe-middleware-for-distributed-ev-charging-optimization", + "title": "Using publish/subscribe middleware for distributed EV charging optimization", + "source_url": "https://doi.org/10.1007/s00450-014-0278-4" + } +] diff --git a/docs/publication-import.md b/docs/publication-import.md new file mode 100644 index 0000000..bed5669 --- /dev/null +++ b/docs/publication-import.md @@ -0,0 +1,91 @@ +# Publication backfill — 2026-09-10 + +Added 336 publication records and reviewed the 36 existing records. The initial archive +contained 372 entries from 2002–2026, including the existing dataset entry. +A subsequent review removed 29 abstract-only/short poster and demo records, +leaving 343 entries (342 papers and one dataset). +The scope is papers coauthored by Hans-Arno Jacobsen and at least one person in +the current or alumni roster. Existing entries outside that scope were retained. + +## Sources and version selection + +- Queried all 486 records in [Jacobsen's DBLP bibliography](https://dblp.org/pid/j/HansArnoJacobsen) + through the [DBLP knowledge graph](https://sparql.dblp.org/). Of these, 390 + paper records matched another member; three matching dataset/software records + were excluded from the paper backfill. +- Compared 567 [OpenAlex records](https://openalex.org/authors/A5072791865) + to find gaps, then checked additions against publisher, conference, arXiv, + or institutional records. Each paper uses `external_url` as its single paper + link. Import sources are recorded separately in + [publication provenance](publication-provenance.json). +- Preserved ordered author lists and existing page paths, notes, and dataset + associations. Added verified author-name variants to nine member profiles. +- Matched 55 arXiv identifiers to published conference, workshop, journal, or + book versions. These share one entry with the published version. Distinct + conference and journal papers remain separate, even when their titles match. +- Seven existing arXiv links now point to published versions. The archive retains + 31 preprints for which a corresponding published version was not confirmed. + A related poster or a similar title alone is insufficient to merge papers. + +## Metadata decisions + +- The [IJCAI survey's proceedings page](https://www.ijcai.org/proceedings/2024/919) + and PDF agree on the title and authors. Crossref returned unrelated metadata + for its DOI, so the entry links directly to IJCAI. +- The [graph partitioning comparison](https://doi.org/10.48786/edbt.2025.14) + belongs to EDBT 2025; DBLP's 2024 year was corrected using DataCite and the + official proceedings. Journal issue years take precedence over early online + dates when an issue year is available. +- The PADRES book chapter appeared twice across the indexes, with different + years and author-name variants. It is represented once using the 2010 DBLP + record and the publisher DOI. +- Alexander Erben/Isenko share DBLP author identifier `314/5970`. + Michalis Bachras publishes as Michail Bachras; Alex Cheung appears as + Alex King Yeung Cheung; Mohammadreza Najafi also appears as Mohammedreza Najafi. + Accents, capitalization, and punctuation variants are retained for matching. + +## Coverage limits + +This is a public-source backfill, not a guarantee that every historical paper +has been indexed. Unmatched member names do not establish that a member has no +publications. No abstracts, summaries, or missing biographical details were generated. + +Edited proceedings, software/data deposits, and presentations without a paper +were not imported as papers. Institutional records identify “The World Cup of +Event Processing” as a summer-school presentation and “SDN-like: a +network-as-a-service publish/subscribe model” as a workshop presentation. +The need- and willingness-based EV charging item is a presentation. The distinct +SDN-like arXiv paper is included. + +“Efficient Filtering of RSS Documents on Computer Cluster” remains unresolved: +the secondary index points to an unavailable CiteSeerX copy and does not establish +a venue. An authoritative group BibTeX export would help close this and any +unindexed gaps. + + +## Abstracts and archive pagination + +- Added 304 original abstracts. Most come from + [OpenAlex API metadata](https://help.openalex.org/api/), published under CC0; + additional text comes from explicitly licensed publisher and repository copies. + Those entries also retain the license link. Abstract sources are recorded in + [publication provenance](publication-provenance.json), rather than in the + publication editor or displayed below abstracts. Published abstracts are + preferred over corresponding preprint abstracts. +- Removed generated summaries from publication metadata, page notes, and the + publication editor. Imported plain text is escaped for Markdown/HTML safety; + whitespace, extraction artifacts, and detached indexing headings are normalized. +- Replaced the RSC graphical-abstract caption with the actual published abstract. + Rejected a PADRES record that ends mid-sentence. The remaining 38 papers need + complete source text with suitable reuse terms; see + [the pending list](publication-abstracts-pending.json). Authors can paste their + original abstracts into the editor. Missing abstracts retain a source link. +- [Excluded records](publication-exclusions.json) include conference abstracts, + extended abstracts, short poster/demo abstracts, a tutorial abstract, and a + workshop announcement. Full papers remain, including longer workshop/demo + papers and the full journal article corresponding to the ICDE presentation. +- The archive renders at most 100 cards per page, including without JavaScript. + Search loads a separate index on first use, caches it in memory, and renders only + the current 100 matching records. Typing is debounced by 150 ms. The index covers + titles, author aliases, venues, years, and abstracts across every page. +- Paper and license links open in a new tab with `noopener noreferrer`. diff --git a/layouts/partials/publication-card.html b/layouts/partials/publication-card.html new file mode 100644 index 0000000..11cae88 --- /dev/null +++ b/layouts/partials/publication-card.html @@ -0,0 +1,10 @@ +
+

{{ partial "publication-authors.html" (dict "authors" .Params.authors) }}

+

{{ .Title }}

+

{{ .Params.venue }}, {{ .Params.year }} · {{ .Params.publication_type }}

+ {{ with .Params.tags }} +
+ {{ range . }}{{ . }}{{ end }} +
+ {{ end }} +
diff --git a/layouts/partials/publication-pagination.html b/layouts/partials/publication-pagination.html new file mode 100644 index 0000000..f0cb642 --- /dev/null +++ b/layouts/partials/publication-pagination.html @@ -0,0 +1,6 @@ +{{ $pager := .pager }} + diff --git a/layouts/publications/archive.html b/layouts/publications/archive.html index 1d7be8e..760714e 100644 --- a/layouts/publications/archive.html +++ b/layouts/publications/archive.html @@ -1,5 +1,6 @@ {{ define "main" }} {{ $pages := sort .RegularPages "Params.year" "desc" }} + {{ $pager := .Paginate $pages 100 }}

{{ .Title }}

@@ -7,9 +8,9 @@

{{ .Title }}

-
+