verl icon indicating copy to clipboard operation
verl copied to clipboard

[fsdp, doc] feat: Enhance FSDP checkpointing with DCP support

Open dkremez opened this issue 1 month ago • 1 comments

This commit introduces support for PyTorch Distributed Checkpoint (DCP) format in FSDP checkpointing, improving performance and reliability. Key changes include:

  • Updated documentation to reflect the new DCP format and its benefits.
  • Added methods to check for DCP format and load DCP checkpoints in fsdp_model_merger.py.
  • Enhanced FSDPCheckpointManager to handle DCP saves and loads, including support for asynchronous saving.
  • Updated configuration to indicate that async_save is now supported for both FSDP and Megatron backends.

These changes streamline the checkpointing process and ensure compatibility with the latest PyTorch features.

What does this PR do?

Add concise overview of what this PR aims to achieve or accomplish. Reference related GitHub issues and PRs that help with the review.

Checklist Before Starting

  • [x] Search for similar PRs. Paste at least one query link here: ...
  • [x] Format the PR title as [{modules}] {type}: {description} (This will be checked by the CI)
    • {modules} include fsdp, megatron, sglang, vllm, rollout, trainer, ci, training_utils, recipe, hardware, deployment, ray, worker, single_controller, misc, perf, model, algo, env, tool, ckpt, doc, data
    • If this PR involves multiple modules, separate them with , like [megatron, fsdp, doc]
    • {type} is in feat, fix, refactor, chore, test
    • If this PR breaks any API (CLI arguments, config, function signature, etc.), add [BREAKING] to the beginning of the title.
    • Example: [BREAKING][fsdp, megatron] feat: dynamic batching

Test

For changes that can not be tested by CI (e.g., algorithm implementation, new model support), validate by experiment(s) and show results like training curve plots, evaluation results, etc.

API and Usage Example

Demonstrate how the API changes if any, and provide usage example(s) if possible.

# Add code snippet or script demonstrating how to use this

Design & Code Changes

Demonstrate the high-level design if this PR is complex, and list the specific changes.

Checklist Before Submitting

[!IMPORTANT] Please check all the following items before requesting a review, otherwise the reviewer might deprioritize this PR for review.

dkremez avatar Nov 26 '25 15:11 dkremez

CLA assistant check
All committers have signed the CLA.

CLAassistant avatar Nov 26 '25 15:11 CLAassistant