incubator-hama-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Edward J. Yoon" <edwardy...@apache.org>
Subject Fwd: Force split
Date Wed, 29 Apr 2009 09:34:12 GMT
In hbase, There is a force split function. We may want to use this as
a pre-conditioner.

---------- Forwarded message ----------
From: Edward J. Yoon <edwardyoon@apache.org>
Date: Wed, Apr 29, 2009 at 6:17 PM
Subject: Re: Force split
To: hbase-user@hadoop.apache.org


According to my understanding, there is no any code for
MODIFY_TABLE_SPLIT and table.jsp work as below.

Writable[] arr = new Writable[1];
arr[0] = new ImmutableBytesWritable(Bytes.toBytes(key));
master.modifyTable(Bytes.toBytes(tableName),
HConstants.MODIFY_TABLE_SPLIT, arr);

Anyway, I'd like to know how to force split.
Below is the info of 'Regions in randufgse' page.
Rows are sequntial numbers between 0 and 10,000.

If I want to split by "000000000007648" then,

master.modifyTable(Bytes.toBytes(tableName),
HConstants.MODIFY_TABLE_SPLIT, new
ImmutableBytesWritable(Bytes.toBytes("000000000007648"));

Is enough?

d8g055.nhncorp.com:60020:  , 000000000000465
a50891.nhncorp.com:60020: 000000000000465, 000000000000811
a51019.nhncorp.com:60020: 000000000000811, 000000000001134
a51019.nhncorp.com:60020: 000000000001134, 000000000001460
a51010.nhncorp.com:60020: 000000000001460, 000000000001723
a51010.nhncorp.com:60020: 000000000001723, 000000000001988
a51018.nhncorp.com:60020: 000000000001988, 000000000002254
a51018.nhncorp.com:60020: 000000000002254, 000000000002519
a50837.nhncorp.com:60020: 000000000002519, 000000000004648
a50837.nhncorp.com:60020: 000000000004648,

On Wed, Apr 29, 2009 at 5:11 PM, Ryan Rawson <ryanobjc@gmail.com> wrote:
> It works, it has worked in the past...
>
> One thing to remember is you can't always split...  You cant split until all
> the open scanners are closed.  Furthermore you can't split if a region has
> just split.  The reason for this is because once you split a region, files
> aren't rewritten, but the 2 daughter regions hold references to the parent
> region.  Until those regions are resolved via compaction (slow, minutes
> possibly), you can't split those daughter regions.
>
> So check again, maybe your splits are working after all.
>
> If you are doing imports, check out my Randomize.java i posted to the list
> last week.  That helps import performance substantially.
>
> On Wed, Apr 29, 2009 at 12:52 AM, Edward J. Yoon <edwardyoon@apache.org>wrote:
>
>> Hi,
>>
>> I saw that function on the web UI but it seems not implemented yet.
>>
>> public void modifyTable(final byte[] tableName, int op, Writable[] args)
>>  switch (op) {
>>    ...
>>    case MODIFY_TABLE_SPLIT:
>>    case MODIFY_TABLE_COMPACT:
>>    ..
>>  }
>> }
>>
>> Am I right? Pls let me know if i'm wrong.
>> --
>> Best Regards, Edward J. Yoon @ NHN, corp.
>> edwardyoon@apache.org
>> http://blog.udanax.org
>>
>



--
Best Regards, Edward J. Yoon @ NHN, corp.
edwardyoon@apache.org
http://blog.udanax.org



-- 
Best Regards, Edward J. Yoon @ NHN, corp.
edwardyoon@apache.org
http://blog.udanax.org

Mime
View raw message