cc_staff
213
edits
Graham: Difference between revisions
Jump to navigation
Jump to search
clarify
No edit summary |
(clarify) |
||
| Line 171: | Line 171: | ||
'''Important''': You should scale the number of CPUs requested, keeping the ratio of CPUs to GPUs at 3.5 or less on 28 core nodes. For example, if you want to run a job using 4 GPUs, you should request '''at most 14 CPU cores'''. For a job with 1 GPU, you should request '''at most 3 CPU cores'''. Users are allowed to run a few short test jobs (shorter than 1 hour) that break this rule to see how your code performs. | '''Important''': You should scale the number of CPUs requested, keeping the ratio of CPUs to GPUs at 3.5 or less on 28 core nodes. For example, if you want to run a job using 4 GPUs, you should request '''at most 14 CPU cores'''. For a job with 1 GPU, you should request '''at most 3 CPU cores'''. Users are allowed to run a few short test jobs (shorter than 1 hour) that break this rule to see how your code performs. | ||
The two newest Volta nodes have 40 cores so above core requests can be adjusted accordingly. They also have NVLINK, which can provide huge benefits for situation where memory bandwidth between GPUs is the bottleneck. To use one of these NVLINK nodes, it should be requested directly, by adding the option '''--nodelist=gra1337''' or '''--nodelist=gra1338''' to the job submission script. | The two newest Volta nodes have 40 cores so above core requests can be adjusted accordingly, i.e. can use 5 CPU cores per GPU. They also have NVLINK, which can provide huge benefits for situation where memory bandwidth between GPUs is the bottleneck. To use one of these NVLINK nodes, it should be requested directly, by adding the option '''--nodelist=gra1337''' or '''--nodelist=gra1338''' to the job submission script. | ||
<!--T:53--> | <!--T:53--> | ||