[
https://issues.apache.org/jira/browse/MAPREDUCE-3289?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13139640#comment-13139640
]
Todd Lipcon commented on MAPREDUCE-3289:
----------------------------------------
I've tested the MR1 equivalent of this and it makes a significant difference. I
agree we could prefetch other map outputs as well, and generally do a smarter
job of ordering the fetches, but this small patch made a good difference on MR1
- so I don't have any reason to believe it wouldn't help in MR2.
> 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
