v12: Use RESTART_BY_OSERVER=YES for MPT small jobs #778
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
mathomp4
added
the
0 diff
sdrabenh
approved these changes
Nov 24, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Tests at NAS shows that single-node jobs with MPT are very slow with restart writes (sometimes 20 minutes at c24!).
The issue is that MPT has trouble with lots of MPI_GatherV calls in a row on a single node. One solution is to add MPI_Barrier calls but that means going deep into MAPL.
A "simpler" solution (no code change needed) is to just use our "write restart by oserver" functionality which doesn't use MPI_GatherV.
So this PR says "if you are a low-res job and running with MPT, just set
WRITE_RESTART_BY_OSERVER: YES. It's not perfect, but it's faster than fixing up MAPL for now.NOTE: @sshakoor1 this will probably need added to the python setup.