Difference between revisions of "GPU Benchmarks"
Jump to navigation
Jump to search
Line 1: | Line 1: | ||
− | Instructions for Linux: | + | Instructions for running the Gromacs GPU Benchmark Single Node test case on Linux: |
1. Download [https://wiki.scinet.utoronto.ca/wiki/images/a/aa/GROMACS_GPU_BENCH.tar.gz Gromacs GPU Benchmark] | 1. Download [https://wiki.scinet.utoronto.ca/wiki/images/a/aa/GROMACS_GPU_BENCH.tar.gz Gromacs GPU Benchmark] | ||
Line 6: | Line 6: | ||
2. Tar and unzip | 2. Tar and unzip | ||
− | 3. Run build script with either "OPENCL" or "CUDA" | + | 3. Run "build.sh" script with either "OPENCL" or "CUDA" |
− | 4. | + | 4. Run benchmark script run.sh |
5. Report performance for each node type proposed. | 5. Report performance for each node type proposed. |
Revision as of 13:01, 7 April 2017
Instructions for running the Gromacs GPU Benchmark Single Node test case on Linux:
1. Download Gromacs GPU Benchmark
2. Tar and unzip
3. Run "build.sh" script with either "OPENCL" or "CUDA"
4. Run benchmark script run.sh
5. Report performance for each node type proposed.