Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Slurm, fix ntasks per node to one #231

Merged
merged 1 commit into from
Dec 3, 2024
Merged

Conversation

albertz
Copy link
Member

@albertz albertz commented Dec 3, 2024

Fix #229.

(Makes PR #230 obsolete.)

(This seems to be needed since we use an additional explicit srun via PR #212.)

Copy link
Contributor

@michelwi michelwi left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I like this one much better than #230

@albertz
Copy link
Member Author

albertz commented Dec 3, 2024

Note, for PR #230, at least I had some theory why this would be the correct fix, but apparently my theory was wrong.

For this PR here, while I can see that it might fix the issue, and apparently in my tests, it also does fix the issue, I don't fully understand the effect of this, and why exactly it goes wrong without this.

@albertz albertz merged commit a0319f3 into master Dec 3, 2024
4 checks passed
@albertz albertz deleted the albert-fix-tasks-per-node-229 branch December 3, 2024 19:57
@albertz
Copy link
Member Author

albertz commented Dec 3, 2024

I just merged this now, as this is really annoying for me otherwise. But maybe someone can at some later point better explain why this is really the right fix. Or maybe we can just make another followup fix in case we figure out a better way to fix it.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Slurm uneven cpu rqmt causes multiple tasks, i.e. the job being run twice
3 participants