[ 
https://issues.apache.org/jira/browse/YARN-10971?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17426975#comment-17426975
 ] 

Jiangminghai commented on YARN-10971:
-------------------------------------

Thanks for the review [~yehuanhuan]! 
The new patch has been updated.

> Continuous scheduling reduces invalid calculations
> --------------------------------------------------
>
>                 Key: YARN-10971
>                 URL: https://issues.apache.org/jira/browse/YARN-10971
>             Project: Hadoop YARN
>          Issue Type: Improvement
>          Components: fairscheduler
>    Affects Versions: 2.7.2
>            Reporter: Jiangminghai
>            Priority: Minor
>             Fix For: 2.7.2
>
>         Attachments: YARN-10971.0.patch, YARN-10971.1.patch
>
>
> In FairScheduler#continuousSchedulingAttempt(), nodes has been sorted by 
> available resources,iterate all nodes in continuous scheduling. 
>  Should break the loop, when  
> *Resources.fitsIn(minimumAllocation,node.getUnallocatedResource()) == false*.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to