Scaling the memory wall using mixed-precision -- HPG-MxP on an exascale machine
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916844793757696 |
|---|---|
| author | Kashi, Aditya Koukpaizan, Nicholson Lu, Hao Matheson, Michael Oral, Sarp Wang, Feiyi |
| author_facet | Kashi, Aditya Koukpaizan, Nicholson Lu, Hao Matheson, Michael Oral, Sarp Wang, Feiyi |
| contents | Mixed-precision algorithms have been proposed as a way for scientific computing to benefit from some of the gains seen for artificial intelligence (AI) on recent high performance computing (HPC) platforms. A few applications dominated by dense matrix operations have seen substantial speedups by utilizing low precision formats such as FP16. However, a majority of scientific simulation applications are memory bandwidth limited. Beyond preliminary studies, the practical gain from using mixed-precision algorithms on a given HPC system is largely unclear.
The High Performance GMRES Mixed Precision (HPG-MxP) benchmark has been proposed to measure the useful performance of a HPC system on sparse matrix-based mixed-precision applications. In this work, we present a highly optimized implementation of the HPG-MxP benchmark for an exascale system and describe our algorithm enhancements. We show for the first time a speedup of 1.6x using a combination of double- and single-precision on modern GPU-based supercomputers. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_11512 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Scaling the memory wall using mixed-precision -- HPG-MxP on an exascale machine Kashi, Aditya Koukpaizan, Nicholson Lu, Hao Matheson, Michael Oral, Sarp Wang, Feiyi Distributed, Parallel, and Cluster Computing Numerical Analysis Performance 65Y10 G.4; C.4 Mixed-precision algorithms have been proposed as a way for scientific computing to benefit from some of the gains seen for artificial intelligence (AI) on recent high performance computing (HPC) platforms. A few applications dominated by dense matrix operations have seen substantial speedups by utilizing low precision formats such as FP16. However, a majority of scientific simulation applications are memory bandwidth limited. Beyond preliminary studies, the practical gain from using mixed-precision algorithms on a given HPC system is largely unclear. The High Performance GMRES Mixed Precision (HPG-MxP) benchmark has been proposed to measure the useful performance of a HPC system on sparse matrix-based mixed-precision applications. In this work, we present a highly optimized implementation of the HPG-MxP benchmark for an exascale system and describe our algorithm enhancements. We show for the first time a speedup of 1.6x using a combination of double- and single-precision on modern GPU-based supercomputers. |
| title | Scaling the memory wall using mixed-precision -- HPG-MxP on an exascale machine |
| topic | Distributed, Parallel, and Cluster Computing Numerical Analysis Performance 65Y10 G.4; C.4 |
| url | https://arxiv.org/abs/2507.11512 |