Rachelint commented on code in PR #11943:
URL: https://github.com/apache/datafusion/pull/11943#discussion_r1716358223
##########
datafusion/functions-aggregate-common/src/aggregate/groups_accumulator/bool_op.rs:
##########
@@ -68,11 +70,21 @@ where
fn update_batch(
Review Comment:
Yes, I agree.
Maybe we can modify it when we start to support the blocked mode in the
specific `GroupAccumulator` or `GroupValues` impl. Seems it will be a too large
change, if we change all of them once.
This sketch can support some combinations for incremental development:
- Single GroupValues + single GroupAccumulator
- Blocked GroupValues + single GroupAccumulator
- Blocked GroupValues + blocked GroupAccumulator
I want to make the `block_size` same too if both used `GroupValues` and
`GroupAccumulator` are in blocked mode, otherwise it may be too complicated...
And the `block_size` and if we enable `blocked mode` will be dominated by
`GroupValues`.
But I guess maybe it is better for us to support `blocked GroupValues` +
`single GroupAccumulator` too? Because it can eliminate the cpu cost about
splitting and copying in `blocked GroupValues`, although they still exist in
`single GroupAccumulator`.
--
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-unsubscr...@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscr...@datafusion.apache.org
For additional commands, e-mail: github-h...@datafusion.apache.org
