Alejandro Abdelnur commented on YARN-326:

LGTM, 3 nits:

* the parseResourceConfigValue(String v) should require ###mb and ###vcores 
present in the config, else fail. As currently the config is only a ### (for 
mb), we have to mark this as an incompat change.
* the DRFPoliciy#compare() does not need to do a Math.signum(s1.start - 
s2.start), it can be just s1.start - s2.start.
* the DRFPoliciy#compare() should not use the name of the job to determine 
order, in the unlikely care 2 jobs are started at the same sime, the return 
should be zero. Things should work just fine.
> Add multi-resource scheduling to the fair scheduler
> ---------------------------------------------------
>                 Key: YARN-326
>                 URL: https://issues.apache.org/jira/browse/YARN-326
>             Project: Hadoop YARN
>          Issue Type: New Feature
>          Components: scheduler
>    Affects Versions: 2.0.2-alpha
>            Reporter: Sandy Ryza
>            Assignee: Sandy Ryza
>         Attachments: FairSchedulerDRFDesignDoc-1.pdf, 
> FairSchedulerDRFDesignDoc.pdf, YARN-326-1.patch, YARN-326-1.patch, 
> YARN-326-2.patch, YARN-326-3.patch, YARN-326-4.patch, YARN-326-5.patch, 
> YARN-326-6.patch, YARN-326.patch, YARN-326.patch
> With YARN-2 in, the capacity scheduler has the ability to schedule based on 
> multiple resources, using dominant resource fairness.  The fair scheduler 
> should be able to do multiple resource scheduling as well, also using 
> dominant resource fairness.
> More details to come on how the corner cases with fair scheduler configs such 
> as min and max resources will be handled.

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