[
https://issues.apache.org/jira/browse/HADOOP-55?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Owen O'Malley resolved HADOOP-55.
---------------------------------
Resolution: Duplicate
I think the performance gains over the last 1.5 years have mostly fixed this.
There is always more to do, but...
> map-reduce job overhead is too high
> -----------------------------------
>
> Key: HADOOP-55
> URL: https://issues.apache.org/jira/browse/HADOOP-55
> Project: Hadoop
> Issue Type: Bug
> Components: mapred
> Affects Versions: 0.2.0
> Reporter: Yoram Arnon
> Assignee: Sameer Paranjpye
> Priority: Minor
>
> map-reduce job overhead is too high. Running a null map-reduce job, with no
> input, no intermediate data and no output, on a several-hundred node cluster,
> takes several minutes, rather than several seconds.
> Enhanced distribution, launching and monitoring mechanisms should reduce the
> overhead to levels that match human online patience for short jobs.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
