[
https://issues.apache.org/jira/browse/HIVE-8042?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14129035#comment-14129035
]
Xuefu Zhang commented on HIVE-8042:
-----------------------------------
Actually I'd contend that neither defending bugs nor limiting the amount
concurrency justifies a configuration for every behavior. We program to create
and fix bugs but not defend them. :) To limit the concurrency, we do have
hive.exec.parallel.thread.number dedicated for that.
I'm not particular fussy about this one, but the number of options really
confuses people. I'm worried that we are adding more and more confusion either
consciously or unconsciously.
If we'd think the old behavior has a bug and this is to fix it, we will feel
better by not providing such a flag.
> Optionally allow move tasks to run in parallel
> ----------------------------------------------
>
> Key: HIVE-8042
> URL: https://issues.apache.org/jira/browse/HIVE-8042
> Project: Hive
> Issue Type: Bug
> Reporter: Gunther Hagleitner
> Assignee: Gunther Hagleitner
> Attachments: HIVE-8042.1.patch
>
>
> hive.exec.parallel allows one to run different stages of a query in parallel.
> However that applies only to map-reduce tasks. When using large multi insert
> queries there are many MoveTasks that are all executed in sequence on the
> client. There's no real reason for that - they could be run in parallel as
> well (i.e.: the stage graph captures the dependencies and knows which tasks
> can happen in parallel)
> Proposal here is to create a new config var:
> hive.exec.parallel.movetask
> When set to true it run movetasks in parallel also.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
