hadoop-common-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Raghu Angadi (JIRA)" <j...@apache.org>
Subject [jira] Commented: (HADOOP-4396) sort on 400 nodes is now slower than in 18
Date Thu, 16 Oct 2008 16:07:44 GMT

    [ https://issues.apache.org/jira/browse/HADOOP-4396?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12640206#action_12640206
] 

Raghu Angadi commented on HADOOP-4396:
--------------------------------------


That is very strange. Could you post the patch you used? What is the typical write size otherwise?
With the modified write(), actual  write systemcall to the file ends up being of the size
"io.file.buffer.size".

The performance difference you are seeing the 3 minutes over 29 minutes, right? Even 29 minutes
on 0.18 was due to such stranglers?

> sort on 400 nodes is now slower than in 18
> ------------------------------------------
>
>                 Key: HADOOP-4396
>                 URL: https://issues.apache.org/jira/browse/HADOOP-4396
>             Project: Hadoop Core
>          Issue Type: Bug
>          Components: mapred
>    Affects Versions: 0.19.0
>            Reporter: Jothi Padmanabhan
>            Assignee: Jothi Padmanabhan
>            Priority: Blocker
>             Fix For: 0.19.0
>
>
> Sort on 400 nodes on  hadoop release 18 takes about 29 minutes, but with the 19 branch
takes about 32 minutes. This behavior is consistent.

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


Mime
View raw message