| Zugriffsnummer | 53369 |
| Dokumenttyp | Konferenzartikel |
| Peer Review | mit Peer Review |
| Sprache | Englisch |
| Titel | Towards an HPC cluster digital twin and scheduling framework for improved energy efficiency |
| Autor(in); Institution |
Kammeyer, Alexander; 9.4, Metrologie für die digitale Transformation, PTB-Berlin
Burger, Florian; Q.4, Informationstechnologie, PTB-Berlin
Lübbert, Daniel; Q.4, Informationstechnologie, PTB-Berlin
Wolter, Katinka; Freie Universität Berlin, Berlin, GERMANY
|
| Quelle/Jahr | Proceedings of the 18th Conference on Computer Science and Intelligence Systems:(2023), 265 - 268 |
| Schriftenreihe | Annals of Computer Science and Information Systems: 35 |
| ISSN | 2300-5963 |
| DOI | |
| URL | |
| Verlag | Warsaw: Polish Information Processing Society |
| Konferenzangaben | 18th Conference on Computer Science and Intelligence Systems (FedCSIS 2023), Warszawa, 17-20 September, 2023, Polen |
| Zusammenfassung | Abstract. Demand for compute resources and thus energy demand for HPC are steadily increasing while the energy market transforms to renewable energy and is facing significant price increases. Optimizing energy efficiency of HPC clusters is therefore a major concern. Different possible optimization dimensions are discussed in this paper. This paper presents a digital twin design for analyzing and reducing energy consumption of a real-world HPC system. The digital twin is based on the HPC cluster at PTB. The digital twin receives information from multiple internal and external data sources to cover the different optimization opportunities. The digital twin also consists of a scheduling simulation framework that uses the data from the digital twin and real-world job traces to test the influence of the different parameters on the HPC cluster. |
| Kostenfreier Zugang | Open Access Hybrid |
Zitierung
Kammeyer, A., Burger, F., Lübbert, D., & Wolter, K. (2023). Towards an HPC cluster digital twin and scheduling framework for improved energy efficiency. 18th Conference on Computer Science and Intelligence Systems (FedCSIS 2023), Warszawa, 17-20 September, 2023, Polen. http://dx.doi.org/10.15439/2023F3797