hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jean-Daniel Cryans (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-8517) Stochastic Loadbalancer isn't finding steady state on real clusters
Date Mon, 13 May 2013 23:21:16 GMT

    [ https://issues.apache.org/jira/browse/HBASE-8517?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13656511#comment-13656511
] 

Jean-Daniel Cryans commented on HBASE-8517:
-------------------------------------------

I'm having a bit of a hard time approaching this patch, not having the context and not being
familiar with that part of the code. I can't really comment on the functionality so +1 if
it fixes your issue and passes the tests, but here's a few nits:

 - The forceLog thing is misleading, since if you run at INFO it's still not logging. Maybe
logTraceAtDebug would be better? Also add a comment in the code where you set it to true to
say why you do it, it's not very obvious.
 - Like [~yuzhihong@gmail.com] said, the 600 is kind of coming out of nowhere but you gave
me the reason and it makes sense why you can't just pass maxResults. Extract as a constant
then?
 - The +9 for META's move cost should also be extracted as a constant, it would help with
readability (not related to your patch, I know, you were just moving code).
                
> Stochastic Loadbalancer isn't finding steady state on real clusters
> -------------------------------------------------------------------
>
>                 Key: HBASE-8517
>                 URL: https://issues.apache.org/jira/browse/HBASE-8517
>             Project: HBase
>          Issue Type: Bug
>    Affects Versions: 0.95.0
>            Reporter: Elliott Clark
>            Assignee: Elliott Clark
>         Attachments: HBASE-8517-0.patch, HBASE-8517-1.patch
>
>
> I have a cluster that runs IT tests.  Last night after all tests were done I noticed
that the balancer was thrashing regions around.
> The number of regions on each machine is not correct.
> The balancer seems to value the cost of moving a region way too little.
> {code}
> 2013-05-09 16:34:58,920 DEBUG [IPC Server handler 4 on 60000] org.apache.hadoop.hbase.master.balancer.StochasticLoadBalancer:
Finished computing new load balance plan.  Computation took 5367ms to try 8910 different iterations.
 Found a solution that moves 37 regions; Going from a computed cost of 56.50254222730425 to
a new cost of 11.214035466575254
> 2013-05-09 16:37:48,715 DEBUG [IPC Server handler 7 on 60000] org.apache.hadoop.hbase.master.balancer.StochasticLoadBalancer:
Finished computing new load balance plan.  Computation took 4735ms to try 8910 different iterations.
 Found a solution that moves 38 regions; Going from a computed cost of 56.612624531830996
to a new cost of 11.275763861636982
> 2013-05-09 16:38:11,398 DEBUG [IPC Server handler 6 on 60000] org.apache.hadoop.hbase.master.balancer.StochasticLoadBalancer:
Finished computing new load balance plan.  Computation took 4502ms to try 8910 different iterations.
 Found a solution that moves 39 regions; Going from a computed cost of 56.50048461413552 to
a new cost of 11.225352339003237
> {code}
> Each of those balancer runs were triggered when there was no load on the cluster.

--
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

Mime
View raw message