[ 
https://issues.apache.org/jira/browse/YARN-1004?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13740735#comment-13740735
 ] 

Bikas Saha commented on YARN-1004:
----------------------------------

IMO the intent of yarn.scheduler.max is to be an admin setting that restricts 
how much resource can be given to any one container. Its exposed via public 
YARN API. All schedulers are supposed to enforce the admin value and not 
determine it for themselves. Hence I dont think it should be scheduler specific.
yarn.scheduler.min is scheduler internal logic on how it simplifies the 
bin-packing problem. Both current schedulers use it and its not exposed by any 
YARN API because its of no use to the user. We may split it to be scheduler 
specific but do we really see either scheduler not using it in the foreseeable 
future? Perhaps we are causing more grief than good by splitting them.
increment-allocation-mb is used only in the fair scheduler. lets just rename 
that to be scheduler specific.

                
> yarn.scheduler.minimum|maximum|increment-allocation-mb should be prefixed 
> with the scheduler type 
> --------------------------------------------------------------------------------------------------
>
>                 Key: YARN-1004
>                 URL: https://issues.apache.org/jira/browse/YARN-1004
>             Project: Hadoop YARN
>          Issue Type: Improvement
>          Components: scheduler
>    Affects Versions: 2.1.0-beta
>            Reporter: Sandy Ryza
>            Priority: Blocker
>             Fix For: 2.1.0-beta
>
>         Attachments: YARN-1004.patch
>
>
> As yarn.scheduler.minimum-allocation-mb is now a scheduler-specific 
> configuration, and functions differently for the Fair and Capacity 
> schedulers, it would be less confusing for the config names to include the 
> scheduler names, i.e. yarn.scheduler.fair.minimum-allocation-mb, 
> yarn.scheduler.capacity.minimum-allocation-mb, and 
> yarn.scheduler.fifo.minimum-allocation-mb.
> The same goes for yarn.scheduler.increment-allocation-mb, which only exists 
> for the Fair Scheduler, and yarn.scheduler.maximum-allocation-mb, for 
> consistency.
> If we wish to preserve backwards compatibility, we can deprecate the old 
> configs to the new ones. 

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

Reply via email to