feat: unify parallel dimension abstraction and support sglang DP Attention#170
Draft
zhjc1124 wants to merge 1 commit into
Draft
feat: unify parallel dimension abstraction and support sglang DP Attention#170zhjc1124 wants to merge 1 commit into
zhjc1124 wants to merge 1 commit into
Conversation
zhjc1124
changed the title
zhjc1124
force-pushed
the
support_pp
branch
2 times, most recently
from
ccaecd5 to
6060581
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
This PR refactors the parallel dimension abstraction in FlexKV to be framework-agnostic, enabling clean three-way unification across sglang, vLLM, and TensorRT-LLM and introduces full support for sglang
--enable-dp-attentionmodeParallel Dimension Model
After this PR, FlexKV's unified 5-D parallel abstraction is:
dp_client_id = instance_id × dp_size + dp_rankWorkerKey(dp_client_id, pp_rank)effective_tp_size = cp_size × tp_size