hbase-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Ted Yu <yuzhih...@gmail.com>
Subject Re: Review Request: First cut at a working load balancer
Date Sun, 25 Jul 2010 22:50:23 GMT
Looking at https://issues.apache.org/jira/browse/HBASE-2697, I found another
review request: http://review.hbase.org/r/382/#review476 which is related to
HBASE-2692
I am not sure whether I should comment about load balancing in HBASE-2697

On Sun, Jul 25, 2010 at 1:31 PM, Ted Yu <yuzhihong@gmail.com> wrote:

> Jonathan:
> Can you elaborate on how the following information (input tobalanceCluster)
> is gathered ?
>
> Map<HServerInfo,List<HRegionInfo>> serverInfo
>
> Considering the implication from HBASE-2473, some (pre-created) regions may
> stay empty for some time. Would those regions be included in the above Map ?
>
> Thanks
>
>
> On Wed, Jul 7, 2010 at 4:48 PM, Jonathan Gray <jgray@apache.org> wrote:
>
>>
>> -----------------------------------------------------------
>> This is an automatically generated e-mail. To reply, visit:
>> http://review.hbase.org/r/282/
>> -----------------------------------------------------------
>>
>> Review request for hbase.
>>
>>
>> Summary
>> -------
>>
>> Just putting this up to show people.  It's not done but the
>> balanceCluster() method "works" and has a bunch of test cases.
>>
>>
>> This addresses bug HBASE-2697.
>>    http://issues.apache.org/jira/browse/HBASE-2697
>>
>>
>> Diffs
>> -----
>>
>>  branches/0.90_master_rewrite/src/main/java/org/apache/hadoop/hbase/HServerInfo.java
>> 961064
>>  branches/0.90_master_rewrite/src/main/java/org/apache/hadoop/hbase/master/LoadBalancer.java
>> PRE-CREATION
>>  branches/0.90_master_rewrite/src/test/java/org/apache/hadoop/hbase/master/TestLoadBalancer.java
>> PRE-CREATION
>>
>> Diff: http://review.hbase.org/r/282/diff
>>
>>
>> Testing
>> -------
>>
>> Added TestLoadBalancer.testLoadBalancer_balanceCluster().  If you have any
>> good edge cases let me know!
>>
>>
>> Thanks,
>>
>> Jonathan
>>
>>
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message