[
https://issues.apache.org/jira/browse/ARROW-13451?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17396170#comment-17396170
]
Ben Kietzman commented on ARROW-13451:
--------------------------------------
At the same time, it'd be nice to make the helper functions
{{internal::GetKernels}} etc public and well tested.
See also: https://github.com/apache/arrow/pull/10793#discussion_r685358176
> [C++][Compute] Consider removing ScalarAggregateKernel
> ------------------------------------------------------
>
> Key: ARROW-13451
> URL: https://issues.apache.org/jira/browse/ARROW-13451
> Project: Apache Arrow
> Issue Type: Improvement
> Components: C++
> Reporter: Ben Kietzman
> Priority: Major
> Labels: pull-request-available
> Time Spent: 2h 10m
> Remaining Estimate: 0h
>
> Scalar aggregation does not incur large memory overhead for the associated
> KernelState objects, so maybe it'd be acceptable to remove explicit scalar
> aggregation kernels in favor of reusing grouped aggregation kernels with a
> single group. This would decrease our maintenance burden significantly, and
> if the benchmarks don't show a regression for single-group aggregation then
> there's no reason not to.
> Even if there is a performance regression we could bundle the scalar and
> grouped aggregate kernels in the same compute::Function and decide between
> them in Dispatch*, rather than confusingly defining distinct "sum" and
> "hash_sum" functions
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
