You are viewing a plain text version of this content. The canonical link for it is here.

Posted to mapreduce-issues@hadoop.apache.org by "Matt Foley (JIRA)" <ji...@apache.org> on 2012/08/24 01:22:42 UTC

[jira] [Updated] (MAPREDUCE-4087) [Gridmix] GenerateDistCacheData job of Gridmix can become slow in some cases

     [ https://issues.apache.org/jira/browse/MAPREDUCE-4087?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Matt Foley updated MAPREDUCE-4087:
----------------------------------

    Fix Version/s: 3.0.0
                   1.1.0

Based on @Ravi: 31/Mar/12 09:29 "I just committed this to trunk and branch-1."
Marking this fixed in 3.0.0 and 1.1.0.
                
> [Gridmix] GenerateDistCacheData job of Gridmix can become slow in some cases
> ----------------------------------------------------------------------------
>
>                 Key: MAPREDUCE-4087
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4087
>             Project: Hadoop Map/Reduce
>          Issue Type: Bug
>            Reporter: Ravi Gummadi
>            Assignee: Ravi Gummadi
>             Fix For: 1.1.0, 3.0.0
>
>         Attachments: 4087.patch, 4087.trunk.patch
>
>
> In map() method of GenerateDistCacheData job of Gridmix, val.setSize() is done every time based on the bytes to be written to a distributed cache file. When we try to write data to next distributed cache file in the same map task, the size of random data generated in each iteration can become small based on the particular case. This can make this dist cache data generation slow.

--
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