[
https://issues.apache.org/jira/browse/MAPREDUCE-5821?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13960190#comment-13960190
]
Todd Lipcon commented on MAPREDUCE-5821:
----------------------------------------
The issue is that, if the input buffer doesn't have room for the value, it will
allocate a new array. But, whenever the Merger calls nextRawValue on a disk
file, it always first resets the buffer to {{diskIFileValue.getData()}} which
is empty. So, at the entry to that function, the length is always 0, and the
code path which reallocs is always taken.
> IFile merge allocates new byte array for every value
> ----------------------------------------------------
>
> Key: MAPREDUCE-5821
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-5821
> Project: Hadoop Map/Reduce
> Issue Type: Bug
> Components: performance, task
> Affects Versions: 2.4.1
> Reporter: Todd Lipcon
> Assignee: Todd Lipcon
>
> I wrote a standalone benchmark of the MapOutputBuffer and found that it did a
> lot of allocations during the merge phase. After looking at an allocation
> profile, I found that IFile.Reader.nextRawValue() would always allocate a new
> byte array for every value, so the allocation rate goes way up during the
> merge phase of the mapper. I imagine this also affects the reducer input,
> though I didn't profile that.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
