[Performance] [flash_communication_v1] DeepSeek communication optimization on A2 (reduce_scatter + all_gather) #1034
Conversation
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
ApsarasX
reviewed
Jun 3, 2025
| "DeepseekV3ForCausalLM", | ||
| "vllm_ascend.models.deepseek_v2:CustomDeepseekV3ForCausalLM") | ||
|
|
||
| if soc_version == 223 or soc_version == 224: |
There was a problem hiding this comment.
What does soc v223 or soc v224 mean?
| return _ETP | ||
|
|
||
| def get_wp_group() -> GroupCoordinator: | ||
| assert _WP is not None, ( |
There was a problem hiding this comment.
Move the declaration of _WP to global scope?
76 tasks
|
This pull request has conflicts, please resolve those before we can evaluate the pull request. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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Outdated suggestions cannot be applied.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.
What this PR does / why we need it?
This PR introduces communication optimization for Ascend A2 clusters in DeepSeek training. The key improvements include:
Replaces all-reduce and broadcast operations with reduce-scatter and all-gather when dp_size > 1
Reduces communication volume and improves performance
Uses scattered data for addition operations
Shared expert hidden_states + routing expert hidden_states
Residual + hidden_states
Does this PR introduce any user-facing change?
None.
How was this patch tested?
We've tested it on benchmark, it meets our satisfaction.