Modify async_out_proc + multi-step for trace mode to trigger output processor for step (i) when running step (i+1) #27
+38
−16
Conversation
This file contains 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
|
👋 Hi! Thank you for contributing to the vLLM project. Once the PR is approved and ready to go, your PR reviewer(s) can run CI to test the changes comprehensively before merging. To run CI, PR reviewers can do one of these:
🚀 |
cglagovichTT
approved these changes
Oct 28, 2024
…rocessor for step (i) when running step (i+1) Signed-off-by: Salar Hosseini <[email protected]>
skhorasganiTT
force-pushed
the
skhorasgani/async_out_proc_per_trace
branch
from
b09e9b7
to
fb938d9
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
async_out_proc_per_tracetoTTModelRunner::execute_model(currently always enabled when async_output_proc, multi-step, and trace-mode are on) to trigger vllm output processor for step (i) when running step (i+1) such that output processor overhead is amortized over num_scheduler_steps (set to 10 in inference example).offline_inference_tt.pyandserver_example_tt.pyoffline_inference_tt.pyfrom ~12.4 t/s/u to ~13.3 t/s/u.