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 "Ravi Gummadi (JIRA)" <ji...@apache.org> on 2011/07/25 12:51:10 UTC
[jira] [Created] (MAPREDUCE-2725) Make Gridmix configure job
specific config properties for the simulated jobs
Make Gridmix configure job specific config properties for the simulated jobs
----------------------------------------------------------------------------
Key: MAPREDUCE-2725
URL: https://issues.apache.org/jira/browse/MAPREDUCE-2725
Project: Hadoop Map/Reduce
Issue Type: Improvement
Components: contrib/gridmix
Reporter: Ravi Gummadi
Assignee: Ravi Gummadi
As Rumen gets all the config properties into the trace file now(after MAPREDUCE-2153), Gridmix can set some of the important job-specific config properties obtained from trace to the simulated jobs. This imply that with same amount of input/output data for mapTask/reduceTask, the behavior of different other things like number of spills, sizes of spills, number of intermediate merges, etc would be similar to those were seen in original job(s).
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (MAPREDUCE-2725) Make Gridmix configure job
specific config properties for the simulated jobs
Posted by "Ravi Gummadi (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/MAPREDUCE-2725?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Ravi Gummadi updated MAPREDUCE-2725:
------------------------------------
Attachment: MR2725.patch
Here is a patch that sets all the important job-specific configuration properties for simulated jobs. These properties include io.sort.mb, io.sort.factor.
> Make Gridmix configure job specific config properties for the simulated jobs
> ----------------------------------------------------------------------------
>
> Key: MAPREDUCE-2725
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-2725
> Project: Hadoop Map/Reduce
> Issue Type: Improvement
> Components: contrib/gridmix
> Reporter: Ravi Gummadi
> Assignee: Ravi Gummadi
> Attachments: MR2725.patch
>
>
> As Rumen gets all the config properties into the trace file now(after MAPREDUCE-2153), Gridmix can set some of the important job-specific config properties obtained from trace to the simulated jobs. This imply that with same amount of input/output data for mapTask/reduceTask, the behavior of different other things like number of spills, sizes of spills, number of intermediate merges, etc would be similar to those were seen in original job(s).
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira