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] multi-node backward slowdown #6134

Merged
merged 2 commits into from
Nov 14, 2024
Merged

Conversation

BurkeHulk
Copy link
Contributor

📌 Checklist before creating the PR

  • I have created an issue for this PR for traceability
  • The title follows the standard format: [doc/gemini/tensor/...]: A concise description
  • I have added relevant tags if possible for us to better distinguish different PRs
  • I have installed pre-commit: pip install pre-commit && pre-commit install

🚨 Issue number

fixed #6133

📝 What does this PR do?

Removed redundant memcpy during backward communication.
Now performance is as expected:
backward on 1 node: 1.75 s.
backward on 26 nodes: 1.9 s.

💥 Checklist before requesting a review

  • I have linked my PR to an issue (instruction)
  • My issue clearly describes the problem/feature/proposal, with diagrams/charts/table/code if possible
  • I have performed a self-review of my code
  • I have added thorough tests.
  • I have added docstrings for all the functions/methods I implemented

⭐️ Do you enjoy contributing to Colossal-AI?

  • 🌝 Yes, I do.
  • 🌚 No, I don't.

Tell us more if you don't enjoy contributing to Colossal-AI.

@BurkeHulk BurkeHulk requested a review from a team as a code owner November 14, 2024 05:50
@ver217 ver217 merged commit cc40fe0 into hpcaitech:main Nov 14, 2024
3 of 4 checks passed
flybird11111 pushed a commit to flybird11111/ColossalAI that referenced this pull request Nov 18, 2024
* remove redundant memcpy during backward

* get back record_stream
@BurkeHulk BurkeHulk deleted the fix/backward_comm branch December 17, 2024 09:15
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.

[BUG]: multi-node backward slowdown
2 participants