You are viewing a plain text version of this content. The canonical link for it is here.

Posted to common-dev@hadoop.apache.org by "Mahadev konar (JIRA)" <ji...@apache.org> on 2006/05/23 06:08:30 UTC

[jira] Updated: (HADOOP-205) the job tracker does not schedule enough map on the cluster

     [ http://issues.apache.org/jira/browse/HADOOP-205?page=all ]

Mahadev konar updated HADOOP-205:
---------------------------------

    Attachment: maplaunch.patch

This patch allows map/reduce jobs to be scheduled using the pending load rather than the current load. Also, I have made changes in the JobInProgress to keep a running count of number of running/finished maps/reduces.

> the job tracker does not schedule enough map on the cluster
> -----------------------------------------------------------
>
>          Key: HADOOP-205
>          URL: http://issues.apache.org/jira/browse/HADOOP-205
>      Project: Hadoop
>         Type: Bug

>   Components: mapred
>     Versions: 0.2
>     Reporter: Owen O'Malley
>     Assignee: Owen O'Malley
>      Fix For: 0.4
>  Attachments: maplaunch.patch
>
> The job tracker during my big sort runs only has 80-120 maps running at a time with 2 tasks/node and 179 nodes and a large queue of map tasks. It seems to be caused by the load balancing using the current load rather than the required load to run all of the queued tasks.

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
   http://issues.apache.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see:
   http://www.atlassian.com/software/jira