lance-format / lance-format/lance
Scalar indexes can do CPU-bound work in async functions
Open
Nobody has claimed this yet.
performance
- Dominant language
- Rust
- Stars
- 7.1k
- Forks
- 852
- Avg merge
- 3d 18h
- Merged PRs (30d)
- 272
Description
I see spans in our service that show >200ms of CPU time devoted to ScalarIndexExpr::evaluate. I think the search() methods it calls can potentially be doing CPU-bound work (for merging large bitmaps, for example). We should move more of this work onto dedicated CPU threadpools.
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by locating ScalarIndexExpr::evaluate and tracing the search() methods it calls, especially the paths that merge large bitmaps. Profile those paths in async functions and determine which CPU-bound work should run on dedicated CPU threadpools; done means the relevant CPU work no longer blocks async execution.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- rust
- Domain
- backend, performance
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 35/100