Mailing-List: contact user-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Reply-To: user@hadoop.apache.org
Received-SPF: pass (athena.apache.org: domain of cembree@gmail.com designates
 209.85.216.177 as permitted sender)
MIME-Version: 1.0
Reply-To: chris@embree.us
In-Reply-To: 
 <CAOcnVr1Eh=ADxT=SApmwWMR8_M7Ym1L_ZYLBFvsn+a309TgtMw@mail.gmail.com>
References: 
 <CAAu13zE-sQDfFLpXF0wGdoeLJGuwNrBSQhz4CHqAv2CDHeJN=g@mail.gmail.com>
	<CAOcnVr1Eh=ADxT=SApmwWMR8_M7Ym1L_ZYLBFvsn+a309TgtMw@mail.gmail.com>
Date: Tue, 18 Dec 2012 17:19:09 -0500
Message-ID: 
 <CADokdcNZzLwTxwPur7UF3HS++so7ok+4vzYustZXBbaw21qOmg@mail.gmail.com>
Subject: Re: datanode write forwarding
From: Chris Embree <cembree@gmail.com>
To: user@hadoop.apache.org
Content-Type: multipart/alternative; boundary=00235429ccac081e2d04d127e31f

--00235429ccac081e2d04d127e31f
Content-Type: text/plain; charset=ISO-8859-1

Harsh,

Is that a change from 1.0 code?  Hortonworks explains it a little
differently.

Thanks for the details and pointer to the code.

Chris (another one)
On Dec 18, 2012 5:14 PM, "Harsh J" <harsh@cloudera.com> wrote:

> Hi,
>
> The received write packet is directly socket-written to the next
> node's receiver (async process - we don't wait for it to hit the disk
> there), and then written onto the disk of self().
>
> You may see HDFS classes BlockReceiver and PacketReceiver in trunk for
> more information.
>
> On Wed, Dec 19, 2012 at 2:26 AM, Jay Vyas <jayunit100@gmail.com> wrote:
> > Hi :
> >
> > Do data nodes, when they receive write requests, forward the request to
> the
> > next datanode in the chain after appending to a file? Or rather, do the
> > write and forward asynchronously and at the same time?
> >
> > --
> > Jay Vyas
> > http://jayunit100.blogspot.com
>
>
>
> --
> Harsh J
>

--00235429ccac081e2d04d127e31f
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<p>Harsh,</p>
<p>Is that a change from 1.0 code?=A0 Hortonworks explains it a little diff=
erently.=A0 </p>
<p>Thanks for the details and pointer to the code.</p>
<p>Chris (another one)</p>
<div class=3D"gmail_quote">On Dec 18, 2012 5:14 PM, &quot;Harsh J&quot; &lt=
;<a href=3D"mailto:harsh@cloudera.com">harsh@cloudera.com</a>&gt; wrote:<br=
 type=3D"attribution"><blockquote class=3D"gmail_quote" style=3D"margin:0 0=
 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Hi,<br>
<br>
The received write packet is directly socket-written to the next<br>
node&#39;s receiver (async process - we don&#39;t wait for it to hit the di=
sk<br>
there), and then written onto the disk of self().<br>
<br>
You may see HDFS classes BlockReceiver and PacketReceiver in trunk for<br>
more information.<br>
<br>
On Wed, Dec 19, 2012 at 2:26 AM, Jay Vyas &lt;<a href=3D"mailto:jayunit100@=
gmail.com">jayunit100@gmail.com</a>&gt; wrote:<br>
&gt; Hi :<br>
&gt;<br>
&gt; Do data nodes, when they receive write requests, forward the request t=
o the<br>
&gt; next datanode in the chain after appending to a file? Or rather, do th=
e<br>
&gt; write and forward asynchronously and at the same time?<br>
&gt;<br>
&gt; --<br>
&gt; Jay Vyas<br>
&gt; <a href=3D"http://jayunit100.blogspot.com" target=3D"_blank">http://ja=
yunit100.blogspot.com</a><br>
<br>
<br>
<br>
--<br>
Harsh J<br>
</blockquote></div>

--00235429ccac081e2d04d127e31f--