[
https://issues.apache.org/jira/browse/HADOOP-11794?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15048097#comment-15048097
]
Yongjun Zhang commented on HADOOP-11794:
----------------------------------------
We probably don't need ChunkUniformSizeInputFormat, and just use
UniformSizeInputFormat (When we break large file into chunks, it make the split
more uniform), when a file doesn't need to be break into chunks, there is a
single entry for it in the fileListing, and we make the entry's chunkLength the
same as its file length.
I was thinking that for the initial implementation, we can just change the
CopyCommitter, as I described in last comment, instead of introducing a reducer
stage for distcp.
Welcome to comment. Thanks.
> distcp can copy blocks in parallel
> ----------------------------------
>
> Key: HADOOP-11794
> URL: https://issues.apache.org/jira/browse/HADOOP-11794
> Project: Hadoop Common
> Issue Type: Improvement
> Components: tools/distcp
> Affects Versions: 0.21.0
> Reporter: dhruba borthakur
> Assignee: Yongjun Zhang
> Attachments: MAPREDUCE-2257.patch
>
>
> The minimum unit of work for a distcp task is a file. We have files that are
> greater than 1 TB with a block size of 1 GB. If we use distcp to copy these
> files, the tasks either take a long long long time or finally fails. A better
> way for distcp would be to copy all the source blocks in parallel, and then
> stich the blocks back to files at the destination via the HDFS Concat API
> (HDFS-222)
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
