You are viewing a plain text version of this content. The canonical link for it is here.
Posted to jira@arrow.apache.org by "Eduardo Ponce (Jira)" <ji...@apache.org> on 2021/09/14 19:09:00 UTC
[jira] [Updated] (ARROW-13997) [C++] restore exec node based query
performance
[ https://issues.apache.org/jira/browse/ARROW-13997?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Eduardo Ponce updated ARROW-13997:
----------------------------------
Description:
The R bindings transitioned to using exec nodes for all queries. Since ARROW-13740 was merged, the performance for our queries has been slower. ARROW-13940 corrected the accidental disabling of multi threading, which recovered some of the performance, but not all of it.
The three benchmarks showing the degradation:
https://conbench.ursa.dev/benchmarks/c6d37c1517af405dbb8ebad83eec010c/
https://conbench.ursa.dev/benchmarks/cec58c9c21234ffbb1bab28b7fe409ef/
https://conbench.ursa.dev/benchmarks/99c04f5ef2fe4caf868904699c2f377c/
For 2 of the three, there was a large increase when ARROW-13740 was merged and plateau, and then a smaller step down when ARROW-13940 was merged, but we still haven't reverted to the pre-change mean.
was:
The R bindings transitioned to using exec nodes for all queries. Since ARROW-13740 was merged, the performance for our queries has been slower. ARROW-13940 corrected the accidental disabling of multi threading, which recovered some if the performance, but not all of it.
The three benchmarks showing the degradation:
https://conbench.ursa.dev/benchmarks/c6d37c1517af405dbb8ebad83eec010c/
https://conbench.ursa.dev/benchmarks/cec58c9c21234ffbb1bab28b7fe409ef/
https://conbench.ursa.dev/benchmarks/99c04f5ef2fe4caf868904699c2f377c/
For 2 of the three, there was a large increase when ARROW-13740 was merged and plateau, and then a smaller step down when ARROW-13940 was merged, but we still haven't reverted to the pre-change mean.
> [C++] restore exec node based query performance
> -----------------------------------------------
>
> Key: ARROW-13997
> URL: https://issues.apache.org/jira/browse/ARROW-13997
> Project: Apache Arrow
> Issue Type: Bug
> Components: C++
> Affects Versions: 6.0.0
> Reporter: Jonathan Keane
> Priority: Blocker
>
> The R bindings transitioned to using exec nodes for all queries. Since ARROW-13740 was merged, the performance for our queries has been slower. ARROW-13940 corrected the accidental disabling of multi threading, which recovered some of the performance, but not all of it.
> The three benchmarks showing the degradation:
> https://conbench.ursa.dev/benchmarks/c6d37c1517af405dbb8ebad83eec010c/
> https://conbench.ursa.dev/benchmarks/cec58c9c21234ffbb1bab28b7fe409ef/
> https://conbench.ursa.dev/benchmarks/99c04f5ef2fe4caf868904699c2f377c/
> For 2 of the three, there was a large increase when ARROW-13740 was merged and plateau, and then a smaller step down when ARROW-13940 was merged, but we still haven't reverted to the pre-change mean.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)