[
https://issues.apache.org/jira/browse/MAPREDUCE-279?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13008327#comment-13008327
]
Tom White commented on MAPREDUCE-279:
-------------------------------------
There's a lot to digest here, but here are a couple of quick initial high-level
comments from a packaging and staging perspective.
I wonder if it would be easier not to move the
src/java/org/apache/hadoop/mapred(uce) trees at this stage. MR 2 could just
depend on the MapReduce JAR produced by the ant file, just like it does for
Common. This would make the introduction of the codebase easier. There are some
changes required in the existing classes, but by the look of things they are
fairly minor and by introducing them in situ (in separate JIRAs) we can be sure
they won't break existing users, and the changes would be easier to track.
Alternatively this work could depend on full mavenization (at least of
MapReduce), but that's probably some way off.
MAPREDUCE-1638 is highly relevant for this work, since it aims to split out the
MR API from the implementation. I've got an in-progress patch for this, which
I'll post soon for discussion.
> 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
>
> Attachments: MR-279.patch, MR-279.patch, MR-279.sh,
> MR-279_MR_files_to_move.txt
>
>
> 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
