Mailing-List: contact hbase-user-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Reply-To: hbase-user@hadoop.apache.org
Received-SPF: pass (athena.apache.org: domain of gcjhhu-hbase-user@m.gmane.org
 designates 80.91.229.12 as permitted sender)
To: hbase-user@hadoop.apache.org
From: "Billy Pearson" <sales@pearsonwholesale.com>
Subject: Re: Hbase Map-reduce Scheduler
Date: Wed, 26 Aug 2009 14:01:58 -0500
Lines: 28
Message-ID: <h740r7$77g$1@ger.gmane.org>
References: <73d592f60908200322n40aaaa93ycbacc67c71ee6f25@mail.gmail.com>
Mime-Version: 1.0
Content-Type: text/plain;
	format=flowed;
	charset="iso-8859-1";
	reply-type=original
Content-Transfer-Encoding: 7bit
In-Reply-To: <73d592f60908200322n40aaaa93ycbacc67c71ee6f25@mail.gmail.com>
Sender: news <news@ger.gmane.org>

I thank you are looking for this jira for reduce assignment
https://issues.apache.org/jira/browse/HBASE-1199

Regions are assigned in the map to be ran on the server hosting the region 
but if there is an idle server it will pull the next map task.

Billy


"bharath vissapragada" 
<bharathvissapragada1990@gmail.com> wrote in 
message news:73d592f60908200322n40aaaa93ycbacc67c71ee6f25@mail.gmail.com...
> Hi all ,
>
> Can any one tell me where i can access some docs which gives a good
> explanation about how a map-reduce scheduler in Hbase
> i.e., How Map regions are created (to minimize data flow through network)
> and how the reduce phases are performed so that we can minimize
> the flow of keys and values across machines .. and how these are planned 
> by
> the scheduler . Also can anyone tell me any related classes in the API
> about the same so that we can control these manually .
>
> Expecting some reply.
>
> Thanks in advance
>