[
https://issues.apache.org/jira/browse/MAPREDUCE-1463?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12835046#action_12835046
]
dhruba borthakur commented on MAPREDUCE-1463:
---------------------------------------------
@Arun: are you suggesting that the job submission process first generate the
input splits, then determines if the number of map tasks is smaller than a
certain value, and if so then set mapreduce.job.reduce.slowstart.completedmaps
to zero?
> Reducer should start faster for smaller jobs
> --------------------------------------------
>
> Key: MAPREDUCE-1463
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-1463
> Project: Hadoop Map/Reduce
> Issue Type: Improvement
> Components: jobtracker
> Reporter: Scott Chen
> Assignee: Scott Chen
> Attachments: MAPREDUCE-1463-v1.patch, MAPREDUCE-1463-v2.patch,
> MAPREDUCE-1463-v3.patch
>
>
> Our users often complain about the slowness of smaller ad-hoc jobs.
> The overhead to wait for the reducers to start in this case is significant.
> It will be good if we can start the reducer sooner in this case.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
