Mailing-List: contact core-dev-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Reply-To: core-dev@hadoop.apache.org
Message-ID: <771089295.1220592104403.JavaMail.jira@brutus>
Date: Thu, 4 Sep 2008 22:21:44 -0700 (PDT)
From: "Owen O'Malley (JIRA)" <jira@apache.org>
To: core-dev@hadoop.apache.org
Subject: [jira] Updated: (HADOOP-3446) The reduce task should not flush the
 in memory file system before starting the reducer
In-Reply-To: <1115045238.1211824682958.JavaMail.jira@brutus>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


     [ https://issues.apache.org/jira/browse/HADOOP-3446?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Owen O'Malley updated HADOOP-3446:
----------------------------------

    Status: Open  (was: Patch Available)

This looks good, but I think we should define the new parameter mapred.reduce.inmem.percent as a percent of the total heap size rather than a percent of mapred.copy.inmem.percent. 

I'd also change the names to:
mapred.reduce.input.buffer.percent
mapred.shuffle.input.buffer.percent
mapred.shuffle.merge.percent

Other than that, it looks good.

> 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: Chris Douglas
>            Priority: Critical
>             Fix For: 0.19.0
>
>         Attachments: 3446-0.patch, 3446-1.patch, 3446-2.patch, 3446-3.patch, 3446-4.patch, 3446-5.patch
>
>
> 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.