[
https://issues.apache.org/jira/browse/HADOOP-3446?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12610240#action_12610240
]
Owen O'Malley commented on HADOOP-3446:
---------------------------------------
Unfortunately, the patch that I used is now hopelessly out of date from
HADOOP-2095, Someone will try to get a new solution for this into 0.19.
> The reduce task should not flush the in memory file system before starting
> the reducer
> --------------------------------------------------------------------------------------
>
> Key: HADOOP-3446
> URL: https://issues.apache.org/jira/browse/HADOOP-3446
> Project: Hadoop Core
> Issue Type: Improvement
> Components: mapred
> Reporter: Owen O'Malley
> Assignee: Owen O'Malley
> Priority: Critical
>
> In the case where the entire reduce inputs fit in ram, we currently force the
> input to disk and re-read it before giving it to the reducer. It would be
> much better if we merged from the ramfs and any spills to feed the reducer
> its input.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
