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

Posted to mapreduce-issues@hadoop.apache.org by "dhruba borthakur (JIRA)" <ji...@apache.org> on 2010/10/07 07:28:33 UTC

[jira] Commented: (MAPREDUCE-2102) Collect iops per task

    [ https://issues.apache.org/jira/browse/MAPREDUCE-2102?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12918789#action_12918789 ] 

dhruba borthakur commented on MAPREDUCE-2102:
---------------------------------------------

The map task typically reads an HDFS file ad writes to local disk. So, this should encompass both reads and writes as well, isn't it?

also, maybe it is more interesting to collect bytes/sec  rather than IOPS. 

> Collect iops per task
> ---------------------
>
>                 Key: MAPREDUCE-2102
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-2102
>             Project: Hadoop Map/Reduce
>          Issue Type: New Feature
>          Components: task
>            Reporter: Arun C Murthy
>             Fix For: 0.22.0
>
>
> We should enhance the MR framework to collect iops per task along with CPU/Memory (MAPREDUCE-220).

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.