Mailing-List: contact core-user-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Reply-To: core-user@hadoop.apache.org
Received-SPF: pass (athena.apache.org: domain of klzhao@gmail.com designates
 209.85.132.244 as permitted sender)
DomainKey-Signature: a=rsa-sha1; c=nofws;
        d=gmail.com; s=gamma;
        h=message-id:date:from:to:subject:in-reply-to:mime-version
         :content-type:content-transfer-encoding:content-disposition
         :references;
        b=V6F+Bx2MOZ45dsGmWsgsKg4BEcocOpJbgAuwhMrztZu/A6QiblbM85BadCkQhxsX6a
         vemsbSUJRJiTDuNmE0t/B6zCN1M23OhIz5Tk1fUlUJeF5HN+rdBwmHYZSCBWDVa8W/9d
         DfaRq06fBWVAqQSO+QeaPYDLzZdhboBXcvoc0=
Message-ID: <ec25226e0808061108i796d4714w73627253f0654cf2@mail.gmail.com>
Date: Wed, 6 Aug 2008 11:08:35 -0700
From: Kevin <klzhao@gmail.com>
To: core-user@hadoop.apache.org
Subject: Re: DFS. How to read from a specific datanode
In-Reply-To: <48996F59.7070004@gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=ISO-2022-JP
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
References: <ec25226e0808051859j4d61952ah342b6dd3bd930b44@mail.gmail.com>
	 <48996F59.7070004@gmail.com>

Yes, the namenode is in charge of deciding the proximity by using
DNSToSwitchMapping. On the other hand, I am exploring the possibility
to let the client decide which data node to connect to, since
sometimes network hierarchy is so complex or dynamic that we better
leave it to the client to find out which datanode is nearest.

-Kevin


On Wed, Aug 6, 2008 at 2:31 AM, Samuel Guo <guosijie@gmail.com> wrote:
> Kevin 写道:
>>
>> Hi,
>>
>> This is about dfs only, not to consider mapreduce. It may sound like a
>> strange need, but sometimes I want to read a block from a specific
>> data node which holds a replica. Figuring out which datanodes have the
>> block is easy. But is there an easy way to specify which datanode I
>> want to load from?
>>
>> Best,
>> -Kevin
>>
>
> DFSClient will choose a node that contains a replicas of the block for you.
> The chosen node will be the nearest node to your client node. This method is
> awesome.
> plz let me know why you want to specify the datanode yourself :)
>