ch4/shm: enable CMA by default #7639
Open
+2
−2
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
CMA should be enabled by default because test result show CMA outperform POSIX in mid to large message size. The default CMA threshold is set to 8KB which matches the eager threshold determined by iqueue size.
|
test:mpich/ch4/most |
|
Will merge this after getting Aurora numbers and verify the latency numbers as well. |
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Outdated suggestions cannot be applied.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.
CMA should be enabled by default because test result show CMA outperform POSIX in mid to large message size.
The CMA threshold is set to 8K to match the eager threshold. 8KB message using POSIX pipeline is about 10% faster than using CMA in intra-NUMA case on Cascade Lake (fig 1), but not in all other cases. So setting the CMA threshold to the same as the eager threshold and completely bypass the POSIX pipeline seems a good idea.
Also, CMA seems not affected by across NUMA access and way outperformed POSIX pipeline.
@intel may need to change this if using eager module other than iqueue.
On Intel Xeon Gold 6226R (Cascade Lake)
On AMD, this seems to be a OK setting. Although AMD has this weird drop at 16KB message size in almost all cases.
TODO: add numbers for Aurora.
Pull Request Description
Author Checklist
Particularly focus on why, not what. Reference background, issues, test failures, xfail entries, etc.
Commits are self-contained and do not do two things at once.
Commit message is of the form:
module: short descriptionCommit message explains what's in the commit.
Whitespace checker. Warnings test. Additional tests via comments.
For non-Argonne authors, check contribution agreement.
If necessary, request an explicit comment from your companies PR approval manager.