You are viewing a plain text version of this content. The canonical link for it is here.
Posted to yarn-issues@hadoop.apache.org by "Arun Suresh (JIRA)" <ji...@apache.org> on 2016/05/03 19:03:12 UTC

[jira] [Commented] (YARN-5035) FairScheduler: Adjust maxAssign dynamically when assignMultiple is turned on

    [ https://issues.apache.org/jira/browse/YARN-5035?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15269086#comment-15269086 ] 

Arun Suresh commented on YARN-5035:
-----------------------------------

[~kasha], You think it is possible to generalize this further ?
I was thinking of integrating YARN-4056 with the fair scheduler. My initial thought process was to replace the assignMultiple with the Bundler.. which would be able to fit an optimal number of containers on a node (optimal, because the containers selected will have complementary resource requirements and thus will have better packing)
This way, you don't really have to worry about an upper limit..

Thoughts ?

> FairScheduler: Adjust maxAssign dynamically when assignMultiple is turned on
> ----------------------------------------------------------------------------
>
>                 Key: YARN-5035
>                 URL: https://issues.apache.org/jira/browse/YARN-5035
>             Project: Hadoop YARN
>          Issue Type: Improvement
>          Components: fairscheduler
>    Affects Versions: 2.8.0
>            Reporter: Karthik Kambatla
>            Assignee: Karthik Kambatla
>
> When assignMultiple is turned on, today the number of containers allocated to node within a single heartbeat is determined by maxAssign. Picking a higher value for this prevents the spreading of the workload, and a smaller number leads to slow ramp up. 
> The proposal is to make it dynamic - half of the unallocated resources on the node. 



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

---------------------------------------------------------------------
To unsubscribe, e-mail: yarn-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: yarn-issues-help@hadoop.apache.org