Tahir, Anam ; Alt, Bastian ; Rizk, Amr ; Koeppl, Heinz (2024)
Load Balancing in Compute Clusters With Delayed Feedback.
In: IEEE Transactions on Computers, 2023, 72 (6)
doi: 10.26083/tuprints-00026536
Artikel, Zweitveröffentlichung, Verlagsversion
 | Es ist eine neuere Version dieses Eintrags verfügbar. |
Kurzbeschreibung (Abstract)
Load balancing arises as a fundamental problem, underlying the dimensioning and operation of many computing and communication systems, such as job routing in data center clusters, multipath communication, Big Data and queueing systems. In essence, the decision-making agent maps each arriving job to one of the possibly heterogeneous servers while aiming at an optimization goal such as load balancing, low average delay or low loss rate. One main difficulty in finding optimal load balancing policies here is that the agent only partially observes the impact of its decisions, e.g., through the delayed acknowledgements of the served jobs. In this paper, we provide a partially observable (PO) model that captures the load balancing decisions in parallel buffered systems under limited information of delayed acknowledgements. We present a simulation model for this PO system to find a load balancing policy in real-time using a scalable Monte Carlo tree search algorithm. We numerically show that the resulting policy outperforms other limited information load balancing strategies such as variants of Join-the-Most-Observations and has comparable performance to full information strategies like: Join-the-Shortest-Queue, Join-the-Shortest-Queue(d) and Shortest-Expected-Delay. Finally, we show that our approach can optimise the real-time parallel processing by using network data provided by Kaggle.
| Typ des Eintrags: | Artikel |
|---|---|
| Erschienen: | 2024 |
| Autor(en): | Tahir, Anam ; Alt, Bastian ; Rizk, Amr ; Koeppl, Heinz |
| Art des Eintrags: | Zweitveröffentlichung |
| Titel: | Load Balancing in Compute Clusters With Delayed Feedback |
| Sprache: | Englisch |
| Publikationsjahr: | 30 September 2024 |
| Ort: | Darmstadt |
| Publikationsdatum der Erstveröffentlichung: | 2023 |
| Ort der Erstveröffentlichung: | [Erscheinungsort nicht ermittelbar] |
| Verlag: | IEEE |
| Titel der Zeitschrift, Zeitung oder Schriftenreihe: | IEEE Transactions on Computers |
| Jahrgang/Volume einer Zeitschrift: | 72 |
| (Heft-)Nummer: | 6 |
| Kollation: | 13 Seiten |
| DOI: | 10.26083/tuprints-00026536 |
| URL / URN: | https://tuprints.ulb.tu-darmstadt.de/26536 |
| Zugehörige Links: | |
| Herkunft: | Zweitveröffentlichungsservice |
| Kurzbeschreibung (Abstract): | Load balancing arises as a fundamental problem, underlying the dimensioning and operation of many computing and communication systems, such as job routing in data center clusters, multipath communication, Big Data and queueing systems. In essence, the decision-making agent maps each arriving job to one of the possibly heterogeneous servers while aiming at an optimization goal such as load balancing, low average delay or low loss rate. One main difficulty in finding optimal load balancing policies here is that the agent only partially observes the impact of its decisions, e.g., through the delayed acknowledgements of the served jobs. In this paper, we provide a partially observable (PO) model that captures the load balancing decisions in parallel buffered systems under limited information of delayed acknowledgements. We present a simulation model for this PO system to find a load balancing policy in real-time using a scalable Monte Carlo tree search algorithm. We numerically show that the resulting policy outperforms other limited information load balancing strategies such as variants of Join-the-Most-Observations and has comparable performance to full information strategies like: Join-the-Shortest-Queue, Join-the-Shortest-Queue(d) and Shortest-Expected-Delay. Finally, we show that our approach can optimise the real-time parallel processing by using network data provided by Kaggle. |
| Freie Schlagworte: | Parallel systems, load balancing, partial observability |
| Status: | Verlagsversion |
| URN: | urn:nbn:de:tuda-tuprints-265367 |
| Sachgruppe der Dewey Dezimalklassifikatin (DDC): | 000 Allgemeines, Informatik, Informationswissenschaft > 004 Informatik 600 Technik, Medizin, angewandte Wissenschaften > 621.3 Elektrotechnik, Elektronik |
| Fachbereich(e)/-gebiet(e): | 18 Fachbereich Elektrotechnik und Informationstechnik 18 Fachbereich Elektrotechnik und Informationstechnik > Self-Organizing Systems Lab |
| Hinterlegungsdatum: | 30 Sep 2024 09:48 |
| Letzte Änderung: | 11 Okt 2024 14:41 |
| PPN: | |
| Export: | |
| Suche nach Titel in: | TUfind oder in Google |
Verfügbare Versionen dieses Eintrags
- Load Balancing in Compute Clusters With Delayed Feedback. (deposited 30 Sep 2024 09:48) [Gegenwärtig angezeigt]
 |
Frage zum Eintrag |
Optionen (nur für Redakteure)
 |
Redaktionelle Details anzeigen |
Drucken
Impressum