Create Utils functions for PP #1042
Open
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description
The implementation for Pipeline Parallelism are splitted into the following small PRs.
This PR is to create Util functions/classes for Jax's PP.
jax_parallel_state.pyis a simplified, Jax version of vLLM's parallel_state.py, to transfer intermediate tensors between each process, and also provide rank-control APIs for model loading, etc.jax_intermediate_tensor.pyis a Jax version of vLLM's IntermediateTensor class. It contains the tensor dictionary that needs to be transfer from one PP rank to another.Tests
E2E test has verified the whole PP implementation works properly.
Checklist
Before submitting this PR, please make sure: