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

Flower Integration V2 #3216

Merged
merged 15 commits into from
Feb 12, 2025
Merged

Flower Integration V2 #3216

merged 15 commits into from
Feb 12, 2025

Conversation

yanchengnv
Copy link
Collaborator

Fixes # .

Description

This PR implements Flower Integration Version 2 to use Flower's new architecture.

In Flower's previous architecture, on the server side, there are two processes: the superlink process and the server-app process. Flare is responsible for starting and stopping both processes. On client side, the supernode and client-app are included into a single process, and Flare is responsible fro starting and stopping this process.

In the new Flower architecture, on the server side, there is a superlink process, and Flare is responsible for starting and stopping this process. Once started, Flare issues the "flwr run" command to the superlink to submit a Flower app. In the meanwhile, Flare issues the "flwr ls" command periodically to query the status of the submitted app until the app is finished. On the client side, there is one process called Supernode, and Flare is responsible for starting and stopping this process. Flare does not need to start the server-app or client-app processes - these will be managed by the Flower system.

The overall integration architecture remains the same. The only difference is how to start and stop Flower apps.

Types of changes

  • Non-breaking change (fix or new feature that would not break existing functionality).
  • Breaking change (fix or new feature that would cause existing functionality to change).
  • New tests added to cover the changes.
  • Quick tests passed locally by running ./runtest.sh.
  • In-line docstrings updated.
  • Documentation updated.

@yanchengnv
Copy link
Collaborator Author

/build

Copy link
Collaborator

@YuanTingHsieh YuanTingHsieh left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM, added some comments

@yanchengnv yanchengnv enabled auto-merge (squash) February 12, 2025 15:54
@yanchengnv
Copy link
Collaborator Author

/build

Copy link
Collaborator

@YuanTingHsieh YuanTingHsieh left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks! LGTM

@yanchengnv
Copy link
Collaborator Author

/build

@YuanTingHsieh
Copy link
Collaborator

/build

@yanchengnv yanchengnv merged commit 6068f38 into NVIDIA:main Feb 12, 2025
20 checks passed
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] Flower versions >=1.13 don't work when running as a FLARE job
2 participants