Support paged kv cache for benchmarks #130
Open
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
poweiw
added
KV-Cache Management
|
@yunfeng-scale is this PR still relevant? |
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Outdated suggestions cannot be applied.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.
This supports building a new engine with paged kv cache or loading existing engine built with KV cache. Otherwise there would be tensor name errors.
Listing this as draft since I think there's some issue with memory management in either benchmarks or generation that causing KV cache to be not freed and I'm getting steadily increasing memory usage and eventually OOM.
Edit: converting to ready to review since I think OOM is a separate issue from this pR #283