hadoop-yarn-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Karthik Kambatla (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (YARN-3054) Preempt policy in FairScheduler may cause mapreduce job never finish
Date Fri, 11 Mar 2016 21:52:41 GMT

    [ https://issues.apache.org/jira/browse/YARN-3054?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15191538#comment-15191538
] 

Karthik Kambatla commented on YARN-3054:
----------------------------------------

Thanks [~peng.zhang]. I understand it now. 

The elegant way of handling this would be to have a preemption priority or even a preemption
cost per container, which is different from the priority that is used for allocation. That
is a larger conversation to be had. Let us move this out of this umbrella and look at it for
both schedulers together. 

That said, I would expect MapReduce to realize that pending mappers are blocked on waiting
reducers and resolve this. MAPREDUCE-6302 and co. attempt to fix this, so you shouldn't see
issues with job completion itself. 

> Preempt policy in FairScheduler may cause mapreduce job never finish
> --------------------------------------------------------------------
>
>                 Key: YARN-3054
>                 URL: https://issues.apache.org/jira/browse/YARN-3054
>             Project: Hadoop YARN
>          Issue Type: Sub-task
>          Components: fairscheduler
>    Affects Versions: 2.6.0
>            Reporter: Peng Zhang
>
> Preemption policy is related with schedule policy now. Using comparator of schedule policy
to find preemption candidate cannot guarantee a subset of containers never be preempted. And
this may cause tasks to be preempted periodically before they finish. So job cannot make any
progress. 
> I think preemption in YARN should got below assurance:
> 1. Mapreduce jobs can get additional resources when others are idle;
> 2. Mapreduce jobs for one user in one queue can still progress with its min share when
others preempt resources back.
> Maybe always preempt the latest app and container can get this? 



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message