Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

fix(parameter_sync): enable multi-thread all-gather and all-to-all #200

Merged
merged 12 commits into from
Jan 6, 2025

Conversation

haolin-nju
Copy link
Collaborator

In previous implementation, all-gather and all-to-all routed experts with HEP are always invoked in a single thread. This PR enables multi-thread all-gather and all-to-all according to the the number of all-gather groups or all-to-all groups.

chatlearn/runtime/engine.py Outdated Show resolved Hide resolved
@haolin-nju haolin-nju merged commit b9cc669 into main Jan 6, 2025
3 checks passed
@haolin-nju haolin-nju deleted the fix-allgather-slow branch January 6, 2025 02:36
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants