[
https://issues.apache.org/jira/browse/IMPALA-3335?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17205084#comment-17205084
]
ASF subversion and git services commented on IMPALA-3335:
---------------------------------------------------------
Commit c9f8d2564d9df86295080b29f1a3ad34c2342c3d in impala's branch
refs/heads/master from Sahil Takiar
[ https://gitbox.apache.org/repos/asf?p=impala.git;h=c9f8d25 ]
IMPALA-3335: Allow single-node optimization with joins
Enable the single-node optimization for queries with joins.
Testing:
* Ran exhaustive tests
* Looped TPC-DS overnight with NUM_NODES=1 against an impalad
mini-cluster with a single dedicated coordinator
Change-Id: I6b189271630214960ed482cb2b552fba9f246770
Reviewed-on: http://gerrit.cloudera.org:8080/16521
Reviewed-by: Impala Public Jenkins <[email protected]>
Tested-by: Impala Public Jenkins <[email protected]>
> Allow single-node optimization with joins.
> ------------------------------------------
>
> Key: IMPALA-3335
> URL: https://issues.apache.org/jira/browse/IMPALA-3335
> Project: IMPALA
> Issue Type: Improvement
> Components: Frontend
> Affects Versions: Impala 2.5.0
> Reporter: Alexander Behm
> Priority: Minor
> Labels: ramp-up
>
> Now that IMPALA-561 has been fixed, we can remove the workaround that
> disables the our single-node optimization for any plan with joins. See
> MaxRowsProcessedVisitor.java:
> {code}
> } else if (caller instanceof HashJoinNode || caller instanceof
> NestedLoopJoinNode) {
> // Revisit when multiple scan nodes can be executed in a single fragment,
> IMPALA-561
> abort_ = true;
> return;
> }
> {code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
