Benchmark: Difference between revisions
Jump to navigation
Jump to search
(새 문서: == HPC Benchmark == https://www.pugetsystems.com/labs/hpc/nvidia-rtx4090-ml-ai-and-scientific-computing-performance-preliminary-2382/ https://www.pugetsystems.com/labs/hpc/how-to-run-an-optimized-hpl-linpack-benchmark-on-amd-ryzen-threadripper-2990wx-32-core-performance-1291/ https://www.amd.com/en/developer.html == Reference ==) |
|||
| (7 intermediate revisions by the same user not shown) | |||
| Line 1: | Line 1: | ||
== HPC Benchmark == | == HPC Benchmark == | ||
{| class="wikitable" | |||
|+ | |||
!Name | |||
!Description | |||
|- | |||
|HPL-GPU | |||
|A custom benchmark to measure GPU performance on top of HPL packages by enabling CUDA so that the benchmark is able to use GPU. | |||
There are lots of HPL-GPU on internet but no single site does not explain how to compile with background information. So we decide to start a single page [[how to compile HPL-GPU]] for someone who is interrest in. | |||
|- | |||
|HPL 2.3 | |||
|software package that solves a (random) dense linear system in double precision (64 bits) arithmetic on distributed-memory computers. It can thus be regarded as a portable as well as freely available implementation of the High Performance Computing Linpack Benchmark<ref>https://netlib.org/benchmark/hpl/</ref> | |||
High Performance Linpack tag: nvcr.io/nvidia/hpc-benchmarks:21.4-hpl | |||
|- | |||
|HPCG 3.1 | |||
|High Performance Conjugate Gradient solver tag: nvcr.io/nvidia/hpc-benchmarks:21.4-hpcg3.1 | |||
|- | |||
|NAMD 3.0a11 | |||
| Molecular Dynamics tag: nvcr.io/hpc/namd:3.0-alpha11 | |||
|- | |||
|LAMMPS | |||
|Molecular Dynamics tag: nvcr.io/hpc/lammps:patch_4May2022 | |||
|- | |||
|TensorFlow 1.15.5 | |||
|ML/AI framework tag: nvcr.io/nvidia/tensorflow:22.09-tf1-py3 | |||
|- | |||
|PyTorch 1.13.0a0 | |||
|ML/AI framework tag: nvcr.io/nvidia/pytorch:22.09-py3 | |||
|- | |||
|NVIDIA HPC-Benchmarks<ref>https://catalog.ngc.nvidia.com/orgs/nvidia/containers/hpc-benchmarks</ref> | |||
|The <code>hpc-benchmarks:23.5</code> container image is provided with the following packages embedded: | |||
* HPL-NVIDIA 23.5.0 | |||
* HPL-AI-NVIDIA 23.5.0 | |||
* HPCG-NVIDIA 23.5.0 | |||
* NVIDIA HPC-X 2.15 for [[CUDA]] 12.x | |||
* NVIDIA NCCL 2.16.5 for CUDA 12.x | |||
* Intel MKL 2020.4-912 | |||
<nowiki>*</nowiki>NVIDIA's HPL benchmark requires [https://developer.nvidia.com/gdrcopy GRDCopy] installed on the system, <nowiki>https://github.com/NVIDIA/gdrcopy#build-and-installation</nowiki> for more information for install GRDCopy | |||
<nowiki>*</nowiki>GDRCopy supports Quadro- and Tesla-class GPUs only<ref>https://github.com/NVIDIA/gdrcopy/issues/198</ref> | |||
|} | |||
==Reference== | |||
https://www.pugetsystems.com/labs/hpc/nvidia-rtx4090-ml-ai-and-scientific-computing-performance-preliminary-2382/ | https://www.pugetsystems.com/labs/hpc/nvidia-rtx4090-ml-ai-and-scientific-computing-performance-preliminary-2382/ | ||
https://www.pugetsystems.com/labs/hpc/how-to-run-an-optimized-hpl-linpack-benchmark-on-amd-ryzen-threadripper-2990wx-32-core-performance-1291/ | https://www.pugetsystems.com/labs/hpc/how-to-run-an-optimized-hpl-linpack-benchmark-on-amd-ryzen-threadripper-2990wx-32-core-performance-1291/ | ||
https://www.amd.com/en/developer.html | https://www.amd.com/en/developer.html | ||
Latest revision as of 12:43, 28 August 2023
HPC Benchmark
| Name | Description |
|---|---|
| HPL-GPU | A custom benchmark to measure GPU performance on top of HPL packages by enabling CUDA so that the benchmark is able to use GPU.
There are lots of HPL-GPU on internet but no single site does not explain how to compile with background information. So we decide to start a single page how to compile HPL-GPU for someone who is interrest in. |
| HPL 2.3 | software package that solves a (random) dense linear system in double precision (64 bits) arithmetic on distributed-memory computers. It can thus be regarded as a portable as well as freely available implementation of the High Performance Computing Linpack Benchmark[1]
High Performance Linpack tag: nvcr.io/nvidia/hpc-benchmarks:21.4-hpl |
| HPCG 3.1 | High Performance Conjugate Gradient solver tag: nvcr.io/nvidia/hpc-benchmarks:21.4-hpcg3.1 |
| NAMD 3.0a11 | Molecular Dynamics tag: nvcr.io/hpc/namd:3.0-alpha11 |
| LAMMPS | Molecular Dynamics tag: nvcr.io/hpc/lammps:patch_4May2022 |
| TensorFlow 1.15.5 | ML/AI framework tag: nvcr.io/nvidia/tensorflow:22.09-tf1-py3 |
| PyTorch 1.13.0a0 | ML/AI framework tag: nvcr.io/nvidia/pytorch:22.09-py3 |
| NVIDIA HPC-Benchmarks[2] | The hpc-benchmarks:23.5 container image is provided with the following packages embedded:
*NVIDIA's HPL benchmark requires GRDCopy installed on the system, https://github.com/NVIDIA/gdrcopy#build-and-installation for more information for install GRDCopy *GDRCopy supports Quadro- and Tesla-class GPUs only[3] |