[ 
https://issues.apache.org/jira/browse/YARN-4360?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15056432#comment-15056432
 ] 

Carlo Curino commented on YARN-4360:
------------------------------------

Rebasing after YARN-4358 got committed. Note: [~imenache] is working on 
YARN-4359 so some of the ugly "instance of" that you see in this patch are 
going to go away (as he moves the LowCostAligned agents forward). 

> Improve GreedyReservationAgent to support "early" allocations, and 
> performance improvements 
> --------------------------------------------------------------------------------------------
>
>                 Key: YARN-4360
>                 URL: https://issues.apache.org/jira/browse/YARN-4360
>             Project: Hadoop YARN
>          Issue Type: Sub-task
>          Components: capacityscheduler, fairscheduler, resourcemanager
>    Affects Versions: 2.8.0
>            Reporter: Carlo Curino
>            Assignee: Carlo Curino
>         Attachments: YARN-4360.2.patch, YARN-4360.3.patch, YARN-4360.patch
>
>
> The GreedyReservationAgent allocates "as late as possible". Per various 
> conversations, it seems useful to have a mirror behavior that allocates as 
> early as possible. Also in the process we leverage improvements from 
> YARN-4358, and implement an RLE-aware StageAllocatorGreedy(RLE), which 
> significantly speeds up allocation.



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

Reply via email to