Comparing Cross-Platform Performance via Node-to-Node Scaling Studies
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866909844166934528 |
|---|---|
| author | Weiss, Kenneth Stitt, Thomas M. Hawkins, Daryl Pearce, Olga Brink, Stephanie Rieben, Robert N. |
| author_facet | Weiss, Kenneth Stitt, Thomas M. Hawkins, Daryl Pearce, Olga Brink, Stephanie Rieben, Robert N. |
| contents | Due to the increasing diversity of high-performance computing architectures, researchers and practitioners are increasingly interested in comparing a code's performance and scalability across different platforms. However, there is a lack of available guidance on how to actually set up and analyze such cross-platform studies. In this paper, we contend that the natural base unit of computing for such studies is a single compute node on each platform and offer guidance in setting up, running, and analyzing node-to-node scaling studies. We propose templates for presenting scaling results of these studies and provide several case studies highlighting the benefits of this approach. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_12166 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Comparing Cross-Platform Performance via Node-to-Node Scaling Studies Weiss, Kenneth Stitt, Thomas M. Hawkins, Daryl Pearce, Olga Brink, Stephanie Rieben, Robert N. Distributed, Parallel, and Cluster Computing Due to the increasing diversity of high-performance computing architectures, researchers and practitioners are increasingly interested in comparing a code's performance and scalability across different platforms. However, there is a lack of available guidance on how to actually set up and analyze such cross-platform studies. In this paper, we contend that the natural base unit of computing for such studies is a single compute node on each platform and offer guidance in setting up, running, and analyzing node-to-node scaling studies. We propose templates for presenting scaling results of these studies and provide several case studies highlighting the benefits of this approach. |
| title | Comparing Cross-Platform Performance via Node-to-Node Scaling Studies |
| topic | Distributed, Parallel, and Cluster Computing |
| url | https://arxiv.org/abs/2510.12166 |