[ 
https://issues.apache.org/jira/browse/YARN-6610?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16128068#comment-16128068
 ] 

Wangda Tan commented on YARN-6610:
----------------------------------

[~templedf],

bq. My previous comment about performance was assuming that you're testing with 
more than 2 resources. If you're testing with only 2 resources, then I'd be 
surprised to see much of a difference. If that's the case, I can take a closer 
look.
Actually before YARN-6788, trunk is 2x faster than the branch (by using perf 
test added by YARN-6775) with 2 resources, the most expensive piece is 
Collection operations (such as Set look up) and unnecessary 
box/unboxing/instance-initialization. So I think we need to be more careful 
here, new feature is good we can optimize performance of the new feature 
slowly, however we should avoid any obvious performance regression for users 
which are not using the feature.

> DominantResourceCalculator.getResourceAsValue() dominant param is no longer 
> appropriate
> ---------------------------------------------------------------------------------------
>
>                 Key: YARN-6610
>                 URL: https://issues.apache.org/jira/browse/YARN-6610
>             Project: Hadoop YARN
>          Issue Type: Sub-task
>          Components: resourcemanager
>    Affects Versions: YARN-3926
>            Reporter: Daniel Templeton
>            Assignee: Daniel Templeton
>            Priority: Critical
>         Attachments: YARN-6610.001.patch, YARN-6610.YARN-3926.002.patch, 
> YARN-6610.YARN-3926.003.patch, YARN-6610.YARN-3926.004.patch, 
> YARN-6610.YARN-3926.005.patch, YARN-6610.YARN-3926.006.patch
>
>
> The {{dominant}} param assumes there are only two resources, i.e. true means 
> to compare the dominant, and false means to compare the subordinate.  Now 
> that there are _n_ resources, this parameter no longer makes sense.



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)

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

Reply via email to