[ 
https://issues.apache.org/jira/browse/YARN-7374?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16213254#comment-16213254
 ] 

Yufei Gu commented on YARN-7374:
--------------------------------

Thanks for working on this [~templedf]. The patch looks good to me generally. 
Some nits:
- Would you like to publish the performance comparison result?
- remove "* @param n the number of resource types" for method {{compare2()}} 
- Two empty lines before "// A queue is needy for its min share if its dominant 
resource".
- Code would be cleaner if putting method {{compare2}} and its support methods 
to a separated class.
- Maybe a good idea to add comment to indicate how to get non-dominate index, 
or a new method like {{getNonDominateIndex(int dominant) { return 1 - 
dominant}}}.
- Could these code be put into separated method? Since it is invoked several 
times. 
{code}
      if (res == 0) {
        // Apps are tied in fairness ratio. Break the tie by submit time and job
        // name to get a deterministic ordering, which is useful for unit tests.
        res = (int) Math.signum(s1.getStartTime() - s2.getStartTime());

        if (res == 0) {
          res = s1.getName().compareTo(s2.getName());
        }
      }
{code}


> Improve performance of DRF comparisons for resource types in fair scheduler
> ---------------------------------------------------------------------------
>
>                 Key: YARN-7374
>                 URL: https://issues.apache.org/jira/browse/YARN-7374
>             Project: Hadoop YARN
>          Issue Type: Sub-task
>          Components: fairscheduler
>    Affects Versions: 3.1.0
>            Reporter: Daniel Templeton
>            Assignee: Daniel Templeton
>            Priority: Critical
>         Attachments: YARN-7374.001.patch
>
>




--
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