Sharding GPU support #289
Open
Sharding GPU support #289
Conversation
This file contains 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
cmd-ntrf
force-pushed
the
mig
branch
5 times, most recently
from
b919f9b
to
211d65d
Compare
|
@etiennedub Can you rebase this and fix conflict now that MIG PRs have been merged? |
etiennedub
force-pushed
the
sharding_doc
branch
from
252b744
to
17d1ace
Compare
added 2 commits
April 12, 2024 08:53
etiennedub
force-pushed
the
sharding_doc
branch
from
17d1ace
to
c56851f
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
It's based on MIG changes because MIG PR changes a bit the way the GPU are configured.
I added one parameter that set the number of sharding for the whole. The shard are evenly split between GPUs on the node. Initially, I wanted to set the set the shard number per GPU but it was complicated to configure, even more considering the MIG setup.
This PR add a new parameter to each infra to set the "shard" number similarly to the MIG configuration. If we prefer, we could set the shard number from
profile::slurm::basedirectly with the hieradata instead.Related Puppet PR: ComputeCanada/puppet-magic_castle#322