Heterogeneous Memory Pool Tuning
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_ | 1866910957309001728 |
|---|---|
| author | Vaverka, Filip Vysocky, Ondrej Riha, Lubomir |
| author_facet | Vaverka, Filip Vysocky, Ondrej Riha, Lubomir |
| contents | We present a lightweight tool for the analysis and tuning of application data placement in systems with heterogeneous memory pools. The tool allows non-intrusively identifying, analyzing, and controlling the placement of individual allocations of the application. We use the tool to analyze a set of benchmarks running on the Intel Sapphire Rapids platform with both HBM and DDR memory. The paper also contains an analysis of the performance of both memory subsystems in terms of read/write bandwidth and latency. The key part of the analysis is to focus on performance if both subsystems are used together. We show that only about 60% to 75% of the data must be placed in HBM memory to achieve 90% of the potential performance of the platform on those benchmarks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_14294 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Heterogeneous Memory Pool Tuning Vaverka, Filip Vysocky, Ondrej Riha, Lubomir Performance We present a lightweight tool for the analysis and tuning of application data placement in systems with heterogeneous memory pools. The tool allows non-intrusively identifying, analyzing, and controlling the placement of individual allocations of the application. We use the tool to analyze a set of benchmarks running on the Intel Sapphire Rapids platform with both HBM and DDR memory. The paper also contains an analysis of the performance of both memory subsystems in terms of read/write bandwidth and latency. The key part of the analysis is to focus on performance if both subsystems are used together. We show that only about 60% to 75% of the data must be placed in HBM memory to achieve 90% of the potential performance of the platform on those benchmarks. |
| title | Heterogeneous Memory Pool Tuning |
| topic | Performance |
| url | https://arxiv.org/abs/2505.14294 |