[
https://issues.apache.org/jira/browse/MAPREDUCE-279?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12995768#comment-12995768
]
Jeff Hammerbacher commented on MAPREDUCE-279:
---------------------------------------------
Hey Arun,
Wow, thanks for reviving one of my favorite old issues! One question: how much
code does the prototype share with Hadoop MapReduce? My understanding is that
it's mostly new code. If that's the case, have you considered creating a
separate Apache Incubator project for the new two-level scheduler? Mesos, for
example, has similar aims and is going the Apache Incubator route. I am aware
of at least one other effort on this front, and having the projects gestate in
the Incubator rather than as a branch of Hadoop would allow them to release
more regularly while young and would be more in line with the dreams of the
project split (having HDFS and MapReduce developed as separate projects). This
seems like a great opportunity to continue the trend of keeping individual ASF
projects small and focused so that releases require less work and can happen
more regularly. What do you think?
Later,
Jeff
> 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
