You are viewing a plain text version of this content. The canonical link for it is here.
Posted to jira@arrow.apache.org by "Antoine Pitrou (Jira)" <ji...@apache.org> on 2020/10/22 16:44:00 UTC

[jira] [Commented] (ARROW-10026) [C++] Improve kernel performance on small batches

    [ https://issues.apache.org/jira/browse/ARROW-10026?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17219157#comment-17219157 ] 

Antoine Pitrou commented on ARROW-10026:
----------------------------------------

So, on {{ArrayArrayKernel<Add, Int64Type>/32768/100}}, we've got roughly:
* 50% CPU time spent on the kernel execution itself
* 8% on {{FunctionExecutorImpl::BindArgs}}
* 13% on {{ScalarExecutor::SetupPreallocation}} (including 12% for {{FunctionExecutorImpl::PrepareOutput}})
* 3.5% on {{PropagateNulls}}
* 2.75% on {{ExecBatchIterator::Next}}
* 8% on {{~Datum}} (including 5% in the jemalloc deallocator)
* a ton of other smaller things


> [C++] Improve kernel performance on small batches
> -------------------------------------------------
>
>                 Key: ARROW-10026
>                 URL: https://issues.apache.org/jira/browse/ARROW-10026
>             Project: Apache Arrow
>          Issue Type: Task
>          Components: C++
>            Reporter: Antoine Pitrou
>            Priority: Major
>
> It seems that invoking some kernels on smallish batches has quite an overhead:
> {code}
> ArrayArrayKernel<Add, Int32Type>/32768/100                      2860 ns         2859 ns       245195 bytes_per_second=10.6727G/s items_per_second=2.86494G/s null_percent=1 size=32.768k
> ArrayArrayKernel<Add, Int32Type>/32768/0                        2752 ns         2751 ns       249316 bytes_per_second=11.093G/s items_per_second=2.97775G/s null_percent=0 size=32.768k
> ArrayArrayKernel<Add, Int32Type>/524288/100                    18633 ns        18630 ns        36548 bytes_per_second=26.2097G/s items_per_second=7.03561G/s null_percent=1 size=524.288k
> ArrayArrayKernel<Add, Int32Type>/524288/0                      18260 ns        18257 ns        38245 bytes_per_second=26.7451G/s items_per_second=7.17933G/s null_percent=0 size=524.288k
> {code}
> We should investigate and try to lighten the overhead.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)