You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@hive.apache.org by "Jimmy Xiang (JIRA)" <ji...@apache.org> on 2015/01/15 20:30:34 UTC
[jira] [Resolved] (HIVE-9391) Spark doesn't obey setting
hive.multigroupby.singlereducer
[ https://issues.apache.org/jira/browse/HIVE-9391?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Jimmy Xiang resolved HIVE-9391.
-------------------------------
Resolution: Invalid
Chatted with Xuefu. This setting doesn't mean to use just 1 reducer. Closed it.
> Spark doesn't obey setting hive.multigroupby.singlereducer
> ----------------------------------------------------------
>
> Key: HIVE-9391
> URL: https://issues.apache.org/jira/browse/HIVE-9391
> Project: Hive
> Issue Type: Bug
> Reporter: Jimmy Xiang
> Priority: Minor
> Fix For: spark-branch
>
>
> With hive.multigroupby.singlereducer set to true, Spark still generates plan with multiple groupby reducers. For example, in working on HIVE-9378, I found, for test groupby_complex_types_multi_single_reducer.q, we have plan like
> {noformat}
> + Reducer 2 <- Map 6 (GROUP, 2)
> + Reducer 4 <- Map 7 (GROUP, 2)
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)