[
https://issues.apache.org/jira/browse/IMPALA-6984?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17172456#comment-17172456
]
ASF subversion and git services commented on IMPALA-6984:
---------------------------------------------------------
Commit c413f9b558d51de877f497590baf14139ad5cf99 in impala's branch
refs/heads/master from Joe McDonnell
[ https://gitbox.apache.org/repos/asf?p=impala.git;h=c413f9b ]
IMPALA-10047: Revert core piece of IMPALA-6984
Performance testing on TPC-DS found a peformance regression
on short queries due to delayed exec status reports. Further
testing traced this back to IMPALA-6984's behavior of
cancelling backends on EOS. The coordinator log show that
CancelBackends() call intermittently taking 10 seconds due
to timing out in the RPC layer.
As a temporary workaround, this reverts the core part of
IMPALA-6984 that added that CancelBackends() call for EOS.
It leaves the rest of IMPALA-6984 intact, as other code has built
on top of it.
Testing:
- Core job
- Performance tests
Change-Id: Ibf00a56e91f0376eaaa552e3bb4763501bfb49e8
(cherry picked from commit b91f3c0e064d592f3cdf2a2e089ca6546133ba55)
Reviewed-on: http://gerrit.cloudera.org:8080/16288
Reviewed-by: Joe McDonnell <[email protected]>
Tested-by: Impala Public Jenkins <[email protected]>
> Coordinator should cancel backends when returning EOS
> -----------------------------------------------------
>
> Key: IMPALA-6984
> URL: https://issues.apache.org/jira/browse/IMPALA-6984
> Project: IMPALA
> Issue Type: Sub-task
> Components: Backend
> Affects Versions: Impala 3.0
> Reporter: Daniel Hecht
> Assignee: Tim Armstrong
> Priority: Major
> Labels: query-lifecycle
> Fix For: Impala 4.0
>
>
> Currently, the Coordinator waits for backends rather than proactively
> cancelling them in the case of hitting EOS. There's a tangled mess that makes
> it tricky to proactively cancel the backends related to how
> {{Coordinator::ComputeQuerySummary()}} works – we can't update the summary
> until the profiles are no longer changing (which also makes sense given that
> we want the exec summary to be consistent with the final profile). But we
> current tie together the FIS status and the profile, and cancellation of
> backends causes the FIS to return CANCELLED, which then means that the
> remaining FIS on that backend won't produce a final profile.
> With the rework of the protocol for IMPALA-2990 we should make it possible to
> sort this out such that a final profile can be requested regardless of how a
> FIS ends execution.
> This also relates to IMPALA-5783.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
