Mailing-List: contact user-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Reply-To: user@hadoop.apache.org
Received-SPF: pass (nike.apache.org: domain of karim.awara@kaust.edu.sa
 designates 74.125.82.170 as permitted sender)
MIME-Version: 1.0
From: Karim Awara <karim.awara@kaust.edu.sa>
Date: Sun, 3 Nov 2013 08:04:48 +0300
Message-ID: 
 <CAHqL09ppPPf70_ZTJVxD-oz7bgYvH7XqHQ8bXqXejWdnh=GqwQ@mail.gmail.com>
Subject: modify writing policy to HDFS
To: user <user@hadoop.apache.org>
Content-Type: multipart/alternative; boundary=f46d044280ca49317904ea3ebecd

--f46d044280ca49317904ea3ebecd
Content-Type: text/plain; charset=US-ASCII

Hi,

I understand the way file upload happens on HDFS, where the data node asks
the namenode for a pipe (64 MB) for writing chunks of the file to hdfs.

I want to change the source code of HDFS such that the datanode can have
multiple pipes opens in parallel, where i push the data to the pipe based
on it content.

so my question is:

1- Is it possible?  if yes, which classes might be responsible for that?
2- how can i track which classes/functions execute a command (for example
when executing a hdfs put command.. how to trace the function calls between
the namnode and datanode?).


Thanks.

--
Best Regards,
Karim Ahmed Awara

-- 

------------------------------
This message and its contents, including attachments are intended solely 
for the original recipient. If you are not the intended recipient or have 
received this message in error, please notify me immediately and delete 
this message from your computer system. Any unauthorized use or 
distribution is prohibited. Please consider the environment before printing 
this email.

--f46d044280ca49317904ea3ebecd
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div><div>Hi, <br><br></div><div>I understand the way file=
 upload happens on HDFS, where the data node asks the namenode for a pipe (=
64 MB) for writing chunks of the file to hdfs.<br>=A0<br></div>I want to ch=
ange the source code of HDFS such that the datanode can have multiple pipes=
 opens in parallel, where i push the data to the pipe based on it content. =
<br>


<br>so my question is: <br><br>1- Is it possible?=A0 if yes, which classes =
might be responsible for that? <br></div><div>2- how can i track which clas=
ses/functions execute a command (for example when executing a hdfs put comm=
and.. how to trace the function calls between the namnode and datanode?). <=
br>


</div><div><br><br></div>Thanks.<br><div><div><div><br clear=3D"all"><div><=
div><div dir=3D"ltr">--<br>Best Regards,<br>Karim Ahmed Awara</div></div>
</div></div></div></div></div>

<br>
<div><hr></div><font face=3D"Arial" size=3D"1">This message and its content=
s, including attachments are intended solely for the original recipient. If=
 you are not the intended recipient or have received this message in error,=
 please notify me immediately and delete this message from your computer sy=
stem. Any unauthorized use or distribution is prohibited. Please consider t=
he environment before printing this email.</font>
--f46d044280ca49317904ea3ebecd--