[
https://issues.apache.org/jira/browse/HBASE-13103?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14366012#comment-14366012
]
Nick Dimiduk commented on HBASE-13103:
--------------------------------------
Sounds about right! I'd say the plan can come up with a single action to
execute per table, and then sleep again -- merging has impact on availability.
Perhaps it wakes up more frequently than the BalancerChore, maybe every minute
instead of every 5 by default. Should also have some back-pressure mechanism,
so it shouldn't start a new merge if existing merge is gummed up. I don't think
we can do that without HBASE-12439. Maybe it cannot be stateless is that's the
case, or perhaps it can check if any merge operations are in flight for the
target table and if so, go back to sleep (or move onto the next table).
> [ergonomics] add region size balancing as a feature of master
> -------------------------------------------------------------
>
> Key: HBASE-13103
> URL: https://issues.apache.org/jira/browse/HBASE-13103
> Project: HBase
> Issue Type: Brainstorming
> Components: Usability
> Reporter: Nick Dimiduk
>
> Often enough, folks miss-judge split points or otherwise end up with a
> suboptimal number of regions. We should have an automated, reliable way to
> "reshape" or "balance" a table's region boundaries. This would be for tables
> that contain existing data. This might look like:
> {noformat}
> Admin#reshapeTable(TableName, int numSplits);
> {noformat}
> or from the shell:
> {noformat}
> > reshape TABLE, numSplits
> {noformat}
> Better still would be to have a maintenance process, similar to the existing
> Balancer that runs AssignmentManager on an interval, to run the above
> "reshape" operation on an interval. That way, the cluster will automatically
> self-correct toward a desirable state.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
