hadoop-mapreduce-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Hudson (Commented) (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (MAPREDUCE-3399) ContainerLocalizer should request new resources after completing the current one
Date Wed, 28 Dec 2011 12:53:30 GMT

    [ https://issues.apache.org/jira/browse/MAPREDUCE-3399?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13176631#comment-13176631

Hudson commented on MAPREDUCE-3399:

Integrated in Hadoop-Hdfs-trunk #908 (See [https://builds.apache.org/job/Hadoop-Hdfs-trunk/908/])
    MAPREDUCE-3399. Modifying ContainerLocalizer to send a heartbeat to NM immediately after
downloading a resource instead of always waiting for a second. Contributed by Siddarth Seth.

vinodkv : http://svn.apache.org/viewcvs.cgi/?root=Apache-SVN&view=rev&rev=1224970
Files : 
* /hadoop/common/trunk/hadoop-mapreduce-project/CHANGES.txt
* /hadoop/common/trunk/hadoop-mapreduce-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-nodemanager/src/main/java/org/apache/hadoop/yarn/server/nodemanager/containermanager/localizer/ContainerLocalizer.java
* /hadoop/common/trunk/hadoop-mapreduce-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-nodemanager/src/test/java/org/apache/hadoop/yarn/server/nodemanager/containermanager/localizer/TestContainerLocalizer.java

> ContainerLocalizer should request new resources after completing the current one
> --------------------------------------------------------------------------------
>                 Key: MAPREDUCE-3399
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-3399
>             Project: Hadoop Map/Reduce
>          Issue Type: Sub-task
>          Components: mrv2, nodemanager
>    Affects Versions: 0.23.0
>            Reporter: Siddharth Seth
>            Assignee: Siddharth Seth
>            Priority: Blocker
>             Fix For: 0.23.1
>         Attachments: MR3399_v1.txt, MR3399_v2.txt
> Currently, the ContainerLocalizer to NM heartbeats to the NM every second. Not very significant,
but this causes a ~4second delay in jobs (job jar, splits, etc). Instead, it should heartbeat
to ask for additional resources to localize as soon as the previous one is localzied. There's
already a TODO in the ContainerLocalizer for this.

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


View raw message