Skip to content

Conversation

d4l3k
Copy link
Member

@d4l3k d4l3k commented Dec 19, 2024

This adds support to the lighthouse and manager to support multiple "rooms" for quorum.

This is intended to be support having a "join" quorum and a "shrink-only" quorum that is needed for LocalSGD (#39)

This may also have the side-effect of being able to support multiple jobs with a single Lighthouse server but that's not being actively targeted currently. We still have shared state in Lighthouse (i.e. heartbeats are global and require globally unique replica_ids).

Test plan:

cargo test
pytest

@facebook-github-bot facebook-github-bot added the CLA Signed This label is managed by the Meta Open Source bot. label Dec 19, 2024
@d4l3k d4l3k force-pushed the d4l3k/multiple_quorum branch from 1982445 to 8f3ec1b Compare December 19, 2024 20:59
@d4l3k d4l3k merged commit 8247faf into main Dec 19, 2024
6 checks passed
@d4l3k d4l3k deleted the d4l3k/multiple_quorum branch December 19, 2024 21:18
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