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

manager: support multiple calls to start_quorum and conditional healing #46

Merged
merged 1 commit into from
Dec 18, 2024

Conversation

d4l3k
Copy link
Member

@d4l3k d4l3k commented Dec 18, 2024

This renames start_step to start_quorum as we can now call it multiple times. The manager step is now only changed during should_commit. We also now support getting a quorum without healing.

allow_heal must be the same on all workers (TBD -- this may be hacky hmm since we don't know what state the workers are in.

This is work towards #39

Test plan:

pytest torchft/manager_test.py torchft/manager_integ_test.py

@facebook-github-bot facebook-github-bot added the CLA Signed This label is managed by the Meta Open Source bot. label Dec 18, 2024
@d4l3k d4l3k merged commit a484e4f into main Dec 18, 2024
6 checks passed
@d4l3k d4l3k deleted the d4l3k/local_sgd_quorum branch December 18, 2024 05:33
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
CLA Signed This label is managed by the Meta Open Source bot.
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants