Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

perf: add v2 fragment file metadata to the FileMetadataCache #2647

Merged
merged 3 commits into from
Jul 27, 2024

Conversation

jiachengdb
Copy link
Contributor

After this cache, we can remove two S3 reads from an ANN query, which can reduce ~100ms query latency.

See #2642

Copy link
Contributor

@westonpace westonpace left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Need to clean up the lint / format changes but this looks good otherwise. I think we eventually want to push the cache down even further. I'd like to cache things like dictionaries, zone maps, etc.

I'd also like to eventually get it to the point where we the cache key is file + column that way we can avoid caching columns that we don't read.

However, this is a good short term win, thanks!

@codecov-commenter
Copy link

Codecov Report

Attention: Patch coverage is 69.04762% with 13 lines in your changes missing coverage. Please review.

Project coverage is 79.38%. Comparing base (a23ba62) to head (cf61172).

Files Patch % Lines
rust/lance-file/src/v2/reader.rs 58.33% 10 Missing ⚠️
rust/lance/src/dataset/fragment.rs 83.33% 0 Missing and 3 partials ⚠️
Additional details and impacted files
@@           Coverage Diff           @@
##             main    #2647   +/-   ##
=======================================
  Coverage   79.38%   79.38%           
=======================================
  Files         221      221           
  Lines       64718    64758   +40     
  Branches    64718    64758   +40     
=======================================
+ Hits        51376    51408   +32     
- Misses      10377    10386    +9     
+ Partials     2965     2964    -1     
Flag Coverage Δ
unittests 79.38% <69.04%> (+<0.01%) ⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

@jiachengdb
Copy link
Contributor Author

@westonpace fixed the lint errors. The remaining test failure is irrelevant.

@eddyxu eddyxu merged commit 8fea797 into lancedb:main Jul 27, 2024
21 of 22 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants