stackSentinel.py Parallel Processing Options #525
Unanswered
phillip-lee-choi
asked this question in
Q&A
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
Hi, stackSentinel.py has the following for parallel processing options:
compute.add_argument('--num_proc', '--num_process', dest='numProcess', type=int, default=1, help='number of tasks running in parallel in each run file (default: %(default)s).')
Can I get some more insight into using this argument with slurm? For instance, if I want to run a job with ntasks=8, then I assume I should add --num_proc=8 as a flag; does --num_proc have to match the exact number of ntasks, or can it be a multiple?
Additionally, what about multithreading? If I use ntasks=4, cpus-per-task=2, then the total number of cores is 8. Should num_process still be 4 to match the number of tasks, or 8 to match the number of cores?
Thanks very much in advance!
Beta Was this translation helpful? Give feedback.
All reactions