[
https://issues.apache.org/jira/browse/ARROW-3120?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17393073#comment-17393073
]
David Li commented on ARROW-3120:
---------------------------------
With the current implementations of ExecNodes, it seems we are leaning towards
the latter for now (e.g. ScalarAggregateNode manually maintains one state per
thread and merges them at the end). That might change if and when we have a
work-stealing scheduler as described on the mailing list.
> [C++] Parallelize execution of ScalarAggregateFunction
> ------------------------------------------------------
>
> Key: ARROW-3120
> URL: https://issues.apache.org/jira/browse/ARROW-3120
> Project: Apache Arrow
> Issue Type: New Feature
> Components: C++
> Reporter: Wes McKinney
> Priority: Major
> Labels: analytics
>
> After ARROW-8972, we have a generic chunk-based executor for aggregate
> functions. It should be relatively straightforward now to parallelize the
> execution loop of the executor.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
