[
https://issues.apache.org/jira/browse/HBASE-5959?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13274435#comment-13274435
]
Elliott Clark commented on HBASE-5959:
--------------------------------------
Move cost is just there to stop thrashing. We want all the other to be way
above what move cost is. If moving were at the same magnitude that
regionSkewCost was then the balancer would not work at all. It would only keep
the random mutations.
Since everything is normalized to between 0 and 1 any multiplier(2 or above)
could work, but 100 just makes things a little bit easier to reason about.
Basically we'll keep a move if it makes the region skew cost better. And then
the balancer will search for how to make that move have the best table skew or
the best data locality .
> Add other load balancers
> ------------------------
>
> Key: HBASE-5959
> URL: https://issues.apache.org/jira/browse/HBASE-5959
> Project: HBase
> Issue Type: New Feature
> Components: master
> Affects Versions: 0.96.0
> Reporter: Elliott Clark
> Assignee: Elliott Clark
> Attachments: HBASE-5959-0.patch, HBASE-5959-1.patch,
> HBASE-5959-2.patch, HBASE-5959-3.patch
>
>
> Now that balancers are pluggable we should give some options.b
--
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
