hbase-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Bryan Duxbury (JIRA)" <j...@apache.org>
Subject [jira] Commented: (HBASE-71) [hbase] Master should rebalance region assignments periodically
Date Fri, 29 Feb 2008 17:58:51 GMT

    [ https://issues.apache.org/jira/browse/HBASE-71?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12573866#action_12573866
] 

Bryan Duxbury commented on HBASE-71:
------------------------------------

How does a region server know when it's overloaded compared to other servers? In DFS we have
the convenient metric of the fixed amount of disk space to tell us when heterogeneous servers
are overloaded. We could do something like memory usage, but that's not necessarily fixed,
nor is it clear that high memory usage actually indicates high load in all circumstances (other
processes may be using a lot of memory, causing swapping, vs HBase being the only program,
so high memory means nothing).

I certainly think that we should be able to manage heterogeneity at some point in the future,
but I think we should put off solving that issue for a while. 

I will go ahead with the plan of region server reporting N regions each time, and if the master
determines the region server's load to be over the average, it will unassign some of those.
Like Stack says, if the region server is still overloaded on the next report in, then it'll
get some more regions shifted around. 

> [hbase] Master should rebalance region assignments periodically
> ---------------------------------------------------------------
>
>                 Key: HBASE-71
>                 URL: https://issues.apache.org/jira/browse/HBASE-71
>             Project: Hadoop HBase
>          Issue Type: Improvement
>          Components: master
>            Reporter: Bryan Duxbury
>            Assignee: Bryan Duxbury
>             Fix For: 0.2.0
>
>
> The master currently only does region assignments at startup or when there are splits
or dead regionservers. This means that if you join a new regionserver to the cluster after
startup, it does not get assigned a fair share of the already-served regions as you would
expect. It only gets a share of new regions being served.
> The master should periodically check the balance of regions, based on whatever assignment
function, instead of in reaction to the above listed events.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


Mime
View raw message