[
https://issues.apache.org/jira/browse/MAPREDUCE-3451?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13159847#comment-13159847
]
Patrick Wendell commented on MAPREDUCE-3451:
--------------------------------------------
Ah, perhaps my issue is a broader misunderstanding of the MR2 architecture.
I thought the Application Master was a long lived service for a particular
application (given its name). I.e. for MapReduce you would expect to have a
single AM running and coordinating submissions for several jobs. Your last
sentence seems to suggest this is incorrect. Is a new AM instantiated for
*each* MR job?
Also, you mentioned priorities. I can't find any details on the scheduling
semantic associated with priorities. I'm assuming those are supposed to convey
the scheduler ordering for requests from a particular AM? I.e. all priority 1
containers should be scheduled before the first priority 2 container? This made
no sense to me before, but it makes more sense if there is a new AM for each
job.
> Port Fair Scheduler to MR2
> --------------------------
>
> Key: MAPREDUCE-3451
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-3451
> Project: Hadoop Map/Reduce
> Issue Type: New Feature
> Components: scheduler
> Reporter: Patrick Wendell
> Assignee: Patrick Wendell
>
> The Fair Scheduler is in widespread use today in MR1 clusters, but not yet
> ported to MR2. This is to track the porting of the Fair Scheduler to MR2 and
> will be updated to include design considerations and progress.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira