GPU-VPC module #3391
Merged
GPU-VPC module #3391
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
cdunbar13
added
the
release-key-new-features
tpdownes
reviewed
Dec 12, 2024
cdunbar13
force-pushed
the
gpu-vpc-promo
branch
from
b117dc2 to
a38915a
Compare
tpdownes
reviewed
Dec 12, 2024
tpdownes
reviewed
Dec 12, 2024
tpdownes
approved these changes
Dec 12, 2024
…new machine-types
…dules/slurm_files/scripts/tools/gpu-test Co-authored-by: Tom Downes <tpdownes@users.noreply.github.com>
cdunbar13
force-pushed
the
gpu-vpc-promo
branch
from
3b40c7a to
7a4d256
Compare
Merged
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Outdated suggestions cannot be applied.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.
This PR adds the
modules/network/gpu-vpcmodule, which is a modified version of therdma-vpcfound in the experimental branch.A number of features have been stripped (ssh and iap firewall, secondary ranges, etc.) and what is left is a module that creates a VPC and several subnets, with the regular use case of inter-gpu communication (i.e. RDMA).
Tested in a modified A3U blueprint. NCCL tests had expected results at 8GB message size.