[
https://issues.apache.org/jira/browse/MAPREDUCE-279?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13002679#comment-13002679
]
Arun C Murthy commented on MAPREDUCE-279:
-----------------------------------------
bq. How shuffle works in MapReduce 2.0 ?
Meng - pretty much the same as currently, the map-outputs are served over http.
We have discussed improvements to shuffle along the lines you have suggested
for a long while now (I just don't have the jiras handy) and I agree, they are
excellent ideas.
Our hope is that with MRv2 we open up Map-Reduce to significant innovation so
that folks can try various ideas like the ones you suggested... make sense?
> Map-Reduce 2.0
> --------------
>
> Key: MAPREDUCE-279
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-279
> Project: Hadoop Map/Reduce
> Issue Type: Improvement
> Components: jobtracker, tasktracker
> Reporter: Arun C Murthy
> Assignee: Arun C Murthy
> Fix For: 0.23.0
>
>
> Re-factor MapReduce into a generic resource scheduler and a per-job,
> user-defined component that manages the application execution.
--
This message is automatically generated by JIRA.
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
