[
https://issues.apache.org/jira/browse/HIVE-5053?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gunther Hagleitner resolved HIVE-5053.
--------------------------------------
Resolution: Won't Fix
Not that important after all. Hive uses stats to automatically assign reducers
through all stages. Also, because of tez having too many reducers isn't really
that critical. And finally one can use the old setting to fix all reducers at
the same time.
> Let user override the parallelism of each tez task
> ---------------------------------------------------
>
> Key: HIVE-5053
> URL: https://issues.apache.org/jira/browse/HIVE-5053
> Project: Hive
> Issue Type: Bug
> Reporter: Gunther Hagleitner
> Fix For: tez-branch
>
>
> We need to come up with a way to let the user choose the parallelism for each
> vertex in the graph. We're numbering the vertices in the graph so we could
> use that to let the user specify the parallelism. Another way would be to
> introduce hints in the sql query itself. But that's a lot more complicated
> for little added value.
--
This message was sent by Atlassian JIRA
(v6.1.4#6159)
