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 "Scott Chen (JIRA)" <ji...@apache.org> on 2011/02/02 21:44:28 UTC
[jira] Commented: (MAPREDUCE-2267) Parallelize reading of blocks
within a stripe
[ https://issues.apache.org/jira/browse/MAPREDUCE-2267?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12989787#comment-12989787 ]
Scott Chen commented on MAPREDUCE-2267:
---------------------------------------
+1 The patch looks good to me
> Parallelize reading of blocks within a stripe
> ---------------------------------------------
>
> Key: MAPREDUCE-2267
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-2267
> Project: Hadoop Map/Reduce
> Issue Type: Improvement
> Components: contrib/raid
> Affects Versions: 0.22.0
> Reporter: Ramkumar Vadali
> Assignee: Ramkumar Vadali
> Attachments: MAPREDUCE-2267.1.patch, MAPREDUCE-2267.2.patch, MAPREDUCE-2267.3.patch, MAPREDUCE-2267.4.patch, MAPREDUCE-2267.patch
>
>
> RAID code has several instances where several blocks of data have to be read to perform an operation. For example, computing a parity block requires reading the blocks of the source file. Similarly, generating a fixed block requires reading a parity block and the good blocks from the source file. These read operations proceed sequentially currently. RAID code should use a thread pool to increase the parallelism and thus reduce latency.
--
This message is automatically generated by JIRA.
-
For more information on JIRA, see: http://www.atlassian.com/software/jira