Vector hybrid search on partitioned table uses brute-force scan despite pre-knn search option being set
Nobody has claimed this yet.
- Dominant language
- C++
- Stars
- 2.9k
- Forks
- 342
- Avg merge
- 2d 14h
- Merged PRs (30d)
- 53
Description
Environment
- Version: 4.6.0 (master branch, pre-release)
- This issue is consistently reproducible.
Description
During vector hybrid search on a partitioned table, the search option parameter specifying pre-knn is not being respected. Instead, the query execution plan defaults to a brute-force scan, which is less efficient for this type of operation.
Impact
The performance of vector hybrid search queries on partitioned tables is degraded, as the intended pre-filtering KNN (K-Nearest Neighbors) optimization is not applied. This leads to unnecessary full scans and slower query response times.
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
Reproduce the vector hybrid search on a partitioned table with the pre-knn search option enabled, then inspect the query execution plan. Done means the option is respected and the plan uses pre-filtering KNN rather than a brute-force scan.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- cpp
- Domain
- databases, performance, search
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Quiet
- Clarity
- Mostly clear
- Newbie friendliness
- 45/100