feat: implemented sparse vector partitioning by lvca · Pull Request #4119 · ArcadeData/arcadedb (original) (raw)

Fixed issue #4087

A query like SELECT vector.neighbors('Doc[embedding]', :query, k) FROM Doc WHERE tenant_id = 'X' on a partitioned('tenant_id') type now narrows the function's per-bucket vector-index fan-out to the partition-derived bucket(s) instead of scanning every bucket. Same for vector.sparseNeighbors. Without a partition-key WHERE the function continues to scan every bucket - no regression.

lvca marked this pull request as ready for review

May 6, 2026 22:24

lvca mentioned this pull request

May 6, 2026

12 tasks

[](/apps/gemini-code-assist)

lvca deleted the vector-partition-pruning branch

May 7, 2026 02:38

tae898 pushed a commit to humemai/arcadedb-embedded-python that referenced this pull request

May 10, 2026

• feat: implemented sparse vector partitioning

Fixed issue ArcadeData#4087

A query like SELECT vector.neighbors('Doc[embedding]', :query, k) FROM Doc WHERE tenant_id = 'X' on a partitioned('tenant_id') type now narrows the function's per-bucket vector-index fan-out to the partition-derived bucket(s) instead of scanning every bucket. Same for vector.sparseNeighbors. Without a partition-key WHERE the function continues to scan every bucket - no regression.

• Added studio support

• Small fixes

robfrank pushed a commit that referenced this pull request

May 12, 2026

• feat: implemented sparse vector partitioning

Fixed issue #4087

A query like SELECT vector.neighbors('Doc[embedding]', :query, k) FROM Doc WHERE tenant_id = 'X' on a partitioned('tenant_id') type now narrows the function's per-bucket vector-index fan-out to the partition-derived bucket(s) instead of scanning every bucket. Same for vector.sparseNeighbors. Without a partition-key WHERE the function continues to scan every bucket - no regression.

• Added studio support

• Small fixes

(cherry picked from commit 1dc68da)

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

[ Show hidden characters]({{ revealButtonHref }})