hbase-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Stack <st...@duboce.net>
Subject Re: about hbase load-balancing
Date Thu, 03 Mar 2011 06:50:23 GMT
Patch on its way.  See
https://issues.apache.org/jira/browse/HBASE-3586.  Should be up soon.
St.Ack



On Wed, Mar 2, 2011 at 10:34 PM, Gaojinchao <gaojinchao@huawei.com> wrote:
>
> hbase version: 0.90.0, 0.90.1
>
> My cluster is 5 RS and 1 hmaster , I create table with multiple regions(50 regions per
RS)
> the parallel is improved, but when the one RS is restarted.
> it may be a large number of adjacent regions is reassigned to the restarted one.
> it reduce the parallelism.
> eg:
>
> inital:
>          RS1   RS2    RS3   RS4    RS5
>  Region  [ a]   [a,b]    [b,c]  [c,d]    [d,f]
>         [f ,g]   [g,h]    ....    ...     ....
>
> RS2 restart:
>          RS1   RS2     RS3  RS4     RS5
>  Region  ...     [ a]     ...    ...      ...
>                [b c]
>                [c d]
>                [d,f]
>                 ....
>
> why doesn't it improve the method of balanceCluster ?
> which can get list of regionsToMove using random fashion instead of order fashion
> or add a configure parameter?
>
>

Mime
View raw message