[
https://issues.apache.org/jira/browse/HADOOP-4565?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12651600#action_12651600
]
Enis Soztutar commented on HADOOP-4565:
---------------------------------------
I agree with Joydeep that if this inputFormat is intended for handling large
number of files, then the changes should be done in the MultiFileInputFormat
itself, rather than introducing a new class. However a quick glance to the
patch indicates that this inputformat works for files having more than one
block.
I guess we should better introduce the file[] -> blocks[] -> {hosts[], racks[]}
-> splits[] logic in FileInputFormat and use this logic both in FileInputFormat
and MultiFileInputFormat.
I have opened an issue for a concrete MultiFileInputFormat implementation
(HADOOP-4741), and attached a patch which I developed some time ago.
HADOOP-4741 depends on this issue.
> MultiFileInputSplit can use data locality information to create splits
> ----------------------------------------------------------------------
>
> Key: HADOOP-4565
> URL: https://issues.apache.org/jira/browse/HADOOP-4565
> Project: Hadoop Core
> Issue Type: Improvement
> Components: mapred
> Reporter: dhruba borthakur
> Assignee: dhruba borthakur
> Attachments: CombineMultiFile.patch, CombineMultiFile2.patch,
> CombineMultiFile3.patch
>
>
> The MultiFileInputFormat takes a set of paths and creates splits based on
> file sizes. Each splits contains a few files an each split are roughly equal
> in size. It would be efficient if we can extend this InputFormat to create
> splits such each all the blocks in one split and either node-local or
> rack-local.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
