Mailing-List: contact hdfs-dev-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Reply-To: hdfs-dev@hadoop.apache.org
Received-SPF: pass (athena.apache.org: domain of lsingh.969@gmail.com
 designates 209.85.214.48 as permitted sender)
MIME-Version: 1.0
In-Reply-To: 
 <CAC3a2cM0ANScEJXnOFpKxueDBmNLft+aWA1z1tddw73DD=2Xqw@mail.gmail.com>
References: 
 <CAC3a2cM0ANScEJXnOFpKxueDBmNLft+aWA1z1tddw73DD=2Xqw@mail.gmail.com>
Date: Tue, 29 Nov 2011 09:53:10 -0500
Message-ID: 
 <CAC3a2cPHSCnN72DyunKmJSMkx7+0iH-B=5mAO=kYNK8Us5JoZw@mail.gmail.com>
Subject: Re: Mapping from JobSplit to Blocks on datanodes
From: Lokendra Singh <lsingh.969@gmail.com>
To: hdfs-dev@hadoop.apache.org
Content-Type: multipart/alternative; boundary=000e0ce03ada2ba5b104b2e0c7e7

--000e0ce03ada2ba5b104b2e0c7e7
Content-Type: text/plain; charset=ISO-8859-1

Any pointers please ?
I could see from the source says that it is not trivial to do.
Was just wanted to confirm if there are some API calls I might have missed.

Regards
Lokendra


On Mon, Nov 28, 2011 at 2:17 AM, Lokendra Singh <lsingh.969@gmail.com>wrote:

> Hi,
>
> How can I find which Job Split does a Block on a datanode belong to? Or
> going the other way, if I can find how a Job Split is eventually stored on
> datanodes i.e. the list of Blocks.
> Any pointers are appreciated.
>
> PS: My eventual objective is 'moving data to computation' (for some
> experiments) and hence was taking ideas from Balancer class to move blocks.
>
> Regards
> Lokendra
>

--000e0ce03ada2ba5b104b2e0c7e7--