Update score calculation for CAGRA-Q instance selection #938
+5
−1
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
enp1s0
added
improvement
cjnolet
approved these changes
May 27, 2025
Member
|
/merge |
github-project-automation
bot
moved this from In Progress
to Done
in Vector Search, ML, & Data Mining Release Board
mythrocks
pushed a commit
to mythrocks/cuvs
that referenced
this pull request
Jun 3, 2025
The current CAGRA-Q instance selection criterion is the same as the one for the standard CAGRA, which is not always optimal for CAGRA-Q. This PR updates the criterion for CAGRA-Q to improve the throughput when `team_size=AUTO`. The size (Byte) of each vector is smaller in a dataset compressed with CAGRA-Q compared to an uncompressed one. Because of this, we may be able to improve throughput by using a smaller `team_size`. This PR updates the scoring method for selecting a CAGRA-Q instance to take that into account. Based on my performance tests for SIFT, GloVe, GIST, NYTimes, and OpenAI 5M, the updated scoring method avoided selecting the worst `team_size` values, unlike the current method. Authors: - tsuki (https://github.com/enp1s0) Approvers: - Corey J. Nolet (https://github.com/cjnolet) URL: rapidsai#938
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 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Outdated suggestions cannot be applied.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.
The current CAGRA-Q instance selection criterion is the same as the one for the standard CAGRA, which is not always optimal for CAGRA-Q. This PR updates the criterion for CAGRA-Q to improve the throughput when
team_size=AUTO.The size (Byte) of each vector is smaller in a dataset compressed with CAGRA-Q compared to an uncompressed one. Because of this, we may be able to improve throughput by using a smaller
team_size. This PR updates the scoring method for selecting a CAGRA-Q instance to take that into account. Based on my performance tests for SIFT, GloVe, GIST, NYTimes, and OpenAI 5M, the updated scoring method avoided selecting the worstteam_sizevalues, unlike the current method.