You are viewing a plain text version of this content. The canonical link for it is here.
Posted to github@arrow.apache.org by "haohuaijin (via GitHub)" <gi...@apache.org> on 2026/08/08 08:27:58 UTC
[I] Improve performance of mask-backed RowSelection::and_then [arrow-rs]
haohuaijin opened a new issue, #10592:
URL: https://github.com/apache/arrow-rs/issues/10592
**Is your feature request related to a problem or challenge?**
`and_then_masks` (the mask × mask path of `RowSelection::and_then`, in `selection/algebra.rs`) runs once per predicate per row group during filter pushdown when the selection is mask-backed. It currently processes one set bit at a time via `set_indices()` / per-bit `BooleanBufferBuilder::append`, which is slow for large or dense selections.
**Describe the solution you'd like**
Process the outer mask one 64-bit word at a time: a word with `k` set bits consumes the next `k` bits of `other`, deposited onto its set positions with a portable software bit-deposit loop. Micro-benchmarks show ~4–6x speedup.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@arrow.apache.org.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] Improve performance of mask-backed RowSelection::and_then [arrow-rs]
Posted by "haohuaijin (via GitHub)" <gi...@apache.org>.
haohuaijin closed issue #10592: Improve performance of mask-backed RowSelection::and_then
URL: https://github.com/apache/arrow-rs/issues/10592
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@arrow.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] Improve performance of mask-backed RowSelection::and_then [arrow-rs]
Posted by "haohuaijin (via GitHub)" <gi...@apache.org>.
haohuaijin commented on issue #10592:
URL: https://github.com/apache/arrow-rs/issues/10592#issuecomment-5230254844
as my test, it is not very useful
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@arrow.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] Improve performance of mask-backed RowSelection::and_then [arrow-rs]
Posted by "haohuaijin (via GitHub)" <gi...@apache.org>.
haohuaijin commented on issue #10592:
URL: https://github.com/apache/arrow-rs/issues/10592#issuecomment-5225539188
take
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@arrow.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org