[
https://issues.apache.org/jira/browse/HADOOP-5222?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12672797#action_12672797
]
Chris Douglas commented on HADOOP-5222:
---------------------------------------
The offset should probably be added after the client id, so the DataNode format
remains a superset of the TaskTracker clienttrace. It's probably not useful to
add the offset to the shuffle trace, but it's available if you wanted to be
consistent.
Out of curiosity, how are you using the offsets in your performance analysis?
> Add offset in client trace
> --------------------------
>
> Key: HADOOP-5222
> URL: https://issues.apache.org/jira/browse/HADOOP-5222
> Project: Hadoop Core
> Issue Type: Improvement
> Components: dfs
> Affects Versions: 0.21.0
> Reporter: Lei Xu
> Priority: Minor
> Fix For: 0.21.0
>
> Attachments: clienttrace.patch
>
>
> By adding offset in client trace, the client trace information can provide
> more accurately information about I/O.
> It is useful for performance analyzing.
> Since there is no random write now, the offset of writing is always zero.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
