[Minor] Fix KV cache block size #39
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
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Outdated suggestions cannot be applied.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 PR fixes the KV cache block size to reserve the GPU memory it should have.
When I tried to reproduce "TABLE IV" in the QServe paper, the throughput was somewhat saturated before the point shown in the paper, regardless of the batch size. I found this was due to a miscalculation of the number of KV cache blocks. (The unit of the block size would be 'bytes', not 'bits').
Using 90% of free GPU memory sometimes doesn't even leave room for the hidden dimension, so I reduced it to 75%.