Koskela, T;
Christidi, I;
Giordano, M;
Dubrovska, E;
Quinn, J;
Maynard, C;
Case, D;
... Deakin, T; + view all
(2023)
Principles for Automated and Reproducible Benchmarking.
In:
ACM International Conference Proceeding Series.
(pp. pp. 609-618).
ACM
Preview |
Text
3624062.3624133.pdf - Published Version Download (629kB) | Preview |
Abstract
The diversity in processor technology used by High Performance Computing (HPC) facilities is growing, and so applications must be written in such a way that they can attain high levels of performance across a range of different CPUs, GPUs, and other accelerators. Measuring application performance across this wide range of platforms becomes crucial, but there are significant challenges to do this rigorously, in a time efficient way, whilst assuring results are scientifically meaningful, reproducible, and actionable. This paper presents a methodology for measuring and analysing the performance portability of a parallel application and shares a software framework which combines and extends adopted technologies to provide a usable benchmarking tool. We demonstrate the flexibility and effectiveness of the methodology and benchmarking framework by showcasing a variety of benchmarking case studies which utilise a stable of supercomputing resources at a national scale.
| Type: | Proceedings paper |
|---|---|
| Title: | Principles for Automated and Reproducible Benchmarking |
| Event: | SC-W 2023: Workshops of The International Conference on High Performance Computing, Network, Storage, and Analysis |
| Open access status: | An open access version is available from UCL Discovery |
| DOI: | 10.1145/3624062.3624133 |
| Publisher version: | https://doi.org/10.1145/3624062.3624133 |
| Language: | English |
| Additional information: | This work is licensed under a Creative Commons Attribution-NonCommercial International 4.0 License. |
| UCL classification: | UCL |
| URI: | https://discovery.ucl.ac.uk/id/eprint/10215101 |
Archive Staff Only
 |
View Item |
