feat(issue2): NVSwitch intranode AllReduce switching threshold model - #20
Open
xyaohubery wants to merge 1 commit into
Open
feat(issue2): NVSwitch intranode AllReduce switching threshold model#20xyaohubery wants to merge 1 commit into
xyaohubery wants to merge 1 commit into
Conversation
Quantitative comparison of three intranode AllReduce implementations: - One-Shot (AllGather + Local Reduce) - Two-Shot Multimem/NVLS (NVSwitch hardware reduction) - Two-Shot P2P (Reduce-Scatter + AllGather on GPU SMs) Builds switching threshold model to auto-select optimal algorithm. Key finding: NVLS is always best when NVSwitch is available, saving ~12% of total SMs vs P2P and reducing latency 3-40x. Co-Authored-By: Claude <noreply@anthropic.com>
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Quantitative analysis of three intranode AllReduce implementations with automatic switching threshold model, addressing Issue #2.
Three Algorithms Compared
Key Findings
NVLS saves ~12% of total SMs vs P2P while being 3-40x faster.
Usage
🤖 Generated with Claude Code