[
https://issues.apache.org/jira/browse/ARROW-16138?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17519837#comment-17519837
]
Tobias Zagorni edited comment on ARROW-16138 at 4/8/22 10:57 PM:
-----------------------------------------------------------------
The thread contention in small batch sizes are largely caused by
copying/destructing shared pointers to DataType. Different threads constantly
changing the refcount of the Int64 DataType seems to causes a lot of inter-core
syncronization
!Flamegraph.png|width=891,height=479!
was (Author: JIRAUSER286565):
The thread contention in small batch sizes are largely caused by
copying/destructing pointers to DataType. Different threads constantly changing
the refcount of the Int64 DataType seems to causes a lot of inter-core
syncronization
!Flamegraph.png|width=891,height=479!
> [C++] Improve performance of ExecuteScalarExpression
> ----------------------------------------------------
>
> Key: ARROW-16138
> URL: https://issues.apache.org/jira/browse/ARROW-16138
> Project: Apache Arrow
> Issue Type: Improvement
> Components: C++
> Reporter: Weston Pace
> Assignee: Tobias Zagorni
> Priority: Major
> Attachments: Flamegraph.png
>
>
> One of the things we want to be able to do in the streaming execution engine
> is process data in small L2 sized batches. Based on literature we might like
> to use batches somewhere in the range of 1k to 16k rows. In ARROW-16014 we
> created a benchmark to measure the performance of ExecuteScalarExpression as
> the size of our batches got smaller. There are two things we observed:
> * Something is causing thread contention. We should be able to get pretty
> close to perfect linear speedup when we are evaluating scalar expressions and
> the batch size fits entirely into L2. We are not seeing that.
> * The overhead of ExecuteScalarExpression is too high when processing small
> batches. Even when the expression is doing real work (e.g. copies,
> comparisons) the execution time starts to be dominated by overhead when we
> have 10k sized batches.
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
