Add VLLM_SCHED_PREFILL_KVC_FREEPCT #89
Open
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
…hen percentage of kv cache is free
gshtras
reviewed
Jul 18, 2024
| @@ -24,6 +24,9 @@ | |||
| ARTIFICIAL_PREEMPTION_MAX_CNT = 500 | |||
|
|
|||
|
|
|||
| VLLM_SCHED_PREFILL_KVC_FREEPCT = float( | |||
| os.getenv("VLLM_SCHED_PREFILL_KVC_FREEPCT", 0.0)) # noqa | |||
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Outdated suggestions cannot be applied.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.
Add VLLM_SCHED_PREFILL_KVC_FREEPCT feature to schedule prefill only when percentage of kv cache is free.
To enable this feature export VLLM_SCHED_PREFILL_KVC_FREEPCT= (float num>0.0).
This helps in large Batch Size offline inference scenarios where prefills can be batched and scheduled when a certain percentage of KV cache is free. If KV cache is below VLLM_SCHED_PREFILL_KVC_FREEPCT, no prefills will be scheduled and decode gets priority. As sequences finish decode, kv cache gets freed up.