[
https://issues.apache.org/jira/browse/HIVE-4936?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13720032#comment-13720032
]
Brock Noland commented on HIVE-4936:
------------------------------------
I implemented a fix for this by rsyncing master \-> slave (thread 0) once and
then executing rsync on the slave (thread 0) \-> slave (thread 1 and 2) and
achieved significant speedup. I will post a patch soon.
> PTest increase performance by parallelizing rsync
> -------------------------------------------------
>
> Key: HIVE-4936
> URL: https://issues.apache.org/jira/browse/HIVE-4936
> Project: Hive
> Issue Type: Bug
> Reporter: Brock Noland
> Assignee: Brock Noland
>
> Currently the bottom neck in PTest2 when adding hosts is the rsync to the
> slaves. Today if we have 3 "threads" which will execute on a slave we rsync
> the source from the master to the slave 3 times. This causes the process to
> be CPU bound on the master.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
