[
https://issues.apache.org/jira/browse/HADOOP-3366?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12596493#action_12596493
]
Sameer Paranjpye commented on HADOOP-3366:
------------------------------------------
To clarify, we use the ramfs in exactly one place. Even there it appears to
hinder a tight implementation rather than help (i.e. we have to special case
DataOutputBuffer for it to avoid value copies, build another workaround to
avoid key copies etc.). Let's not use it.
For in memory merges, an iterator over a key/value sequence in a buffer seems a
much better abstraction than an input stream.
> Shuffle/Merge improvements
> --------------------------
>
> Key: HADOOP-3366
> URL: https://issues.apache.org/jira/browse/HADOOP-3366
> Project: Hadoop Core
> Issue Type: Improvement
> Components: mapred
> Reporter: Arun C Murthy
> Assignee: Arun C Murthy
> Fix For: 0.18.0
>
>
> This is intended to be a meta-issue to track various improvements to
> shuffle/merge in the reducer.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
