cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Sylvain Lebresne (Commented) (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (CASSANDRA-3668) Performance of sstableloader is affected in 1.0.x
Date Fri, 27 Jan 2012 08:40:41 GMT

    [ https://issues.apache.org/jira/browse/CASSANDRA-3668?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13194543#comment-13194543
] 

Sylvain Lebresne commented on CASSANDRA-3668:
---------------------------------------------

bq.  because I don't see where/how it was established that the original problem

To sum up, as as Yuki remarked above, the reason streaming is slower in 1.0 is likely due
to 2 aspects (basically because that's the 2 things that changed):
# We switched from NIO transfer to "standard" read and send. Meaning we possibly adds copies
here and there during transfer.
# We now build indexes (primary and secondary) and bloom filters sequentially with the streaming
while it was previously done in parallel of the streaming.

I don't think we really have established what fraction of the slowdown each is responsible
of. But the idea of adding concurrent streaming is to "fix" the second one.

Yuki, you said last patch improves throughput: but how do we stand with this patch compared
to 0.8?
                
> Performance of sstableloader is affected in 1.0.x
> -------------------------------------------------
>
>                 Key: CASSANDRA-3668
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-3668
>             Project: Cassandra
>          Issue Type: Bug
>          Components: API
>    Affects Versions: 1.0.0
>            Reporter: Manish Zope
>            Assignee: Yuki Morishita
>            Priority: Minor
>              Labels: streaming
>             Fix For: 1.1
>
>         Attachments: 0001-Allow-multiple-connection-in-StreamInSession.patch, 0002-Allow-concurrent-stream-in-StreamOutSession.patch,
0003-Add-threads-option-to-sstableloader.patch, 3688-reply_before_closing_writer.txt, sstable-loader
performance.txt
>
>   Original Estimate: 48h
>  Remaining Estimate: 48h
>
> One of my colleague had reported the bug regarding the degraded performance of the sstable
generator and sstable loader.
> ISSUE :- https://issues.apache.org/jira/browse/CASSANDRA-3589 
> As stated in above issue generator performance is rectified but performance of the sstableloader
is still an issue.
> 3589 is marked as duplicate of 3618.Both issues shows resolved status.But the problem
with sstableloader still exists.
> So opening other issue so that sstbleloader problem should not go unnoticed.
> FYI : We have tested the generator part with the patch given in 3589.Its Working fine.
> Please let us know if you guys require further inputs from our side.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

Mime
View raw message