hadoop-mapreduce-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Sharad Agarwal (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (MAPREDUCE-2405) MR-279: Implement uber-AppMaster (in-cluster LocalJobRunner for MRv2)
Date Mon, 28 Mar 2011 07:38:05 GMT

    [ https://issues.apache.org/jira/browse/MAPREDUCE-2405?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13011941#comment-13011941
] 

Sharad Agarwal commented on MAPREDUCE-2405:
-------------------------------------------

The overall direction looks good. 

Noticed in JobImpl:
{code}

.addTransition(JobState.INITED, JobState.KILL_WAIT,
              JobEventType.JOB_KILL,
              KILL_NEW_JOB_TRANSITION)
{code}

Kill event on INITED state should directly get to KILLED state. Also it should not use KILL_NEW_JOB_TRANSITION
because it does not call the abort logic. Here we need a new Transition which executes abort.



> MR-279: Implement uber-AppMaster (in-cluster LocalJobRunner for MRv2)
> ---------------------------------------------------------------------
>
>                 Key: MAPREDUCE-2405
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-2405
>             Project: Hadoop Map/Reduce
>          Issue Type: Improvement
>          Components: mrv2
>            Reporter: Mahadev konar
>            Assignee: Greg Roelofs
>             Fix For: 0.23.0
>
>         Attachments: MR-2405-MR-1220-yarn.v8.MR-279-hadoop-yarn.patch.txt
>
>
> "Port" MAPREDUCE-1220 to MRv2.  This is an optimization for small jobs wherein all tasks
run on the same node in the same JVM/container.

--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira

Mime
View raw message