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 "Todd Lipcon (Updated) (JIRA)" <ji...@apache.org> on 2011/10/28 02:11:32 UTC

[jira] [Updated] (MAPREDUCE-3289) Make use of fadvise in the NM's shuffle handler

     [ https://issues.apache.org/jira/browse/MAPREDUCE-3289?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Todd Lipcon updated MAPREDUCE-3289:
-----------------------------------

    Attachment: mr-3289.txt

Here's a rather simple implementation that fadvises with WILLNEED just before sending the output, and DONTNEED after sending the output.

We could make it fancier by fadvising "one map output ahead", since in MR2 a single request can fetch multiple outputs, but I'd prefer to leave that for a later JIRA.
                
> Make use of fadvise in the NM's shuffle handler
> -----------------------------------------------
>
>                 Key: MAPREDUCE-3289
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-3289
>             Project: Hadoop Map/Reduce
>          Issue Type: Improvement
>          Components: mrv2, nodemanager, performance
>    Affects Versions: 0.23.0
>            Reporter: Todd Lipcon
>            Assignee: Todd Lipcon
>         Attachments: mr-3289.txt
>
>
> Using the new NativeIO fadvise functions, we can make the NodeManager prefetch map output before it's send over the socket, and drop it out of the fs cache once it's been sent (since it's very rare for an output to have to be re-sent). This improves IO efficiency and reduces cache pollution.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira