Return-Path: 
 <user-return-15302-apmail-hadoop-mapreduce-user-archive=hadoop.apache.org@hadoop.apache.org>
X-Original-To: apmail-hadoop-mapreduce-user-archive@minotaur.apache.org
Delivered-To: apmail-hadoop-mapreduce-user-archive@minotaur.apache.org
Received: from mail.apache.org (hermes.apache.org [140.211.11.3])
	by minotaur.apache.org (Postfix) with SMTP id AB6C411430
	for <apmail-hadoop-mapreduce-user-archive@minotaur.apache.org>;
 Sun, 11 May 2014 03:58:03 +0000 (UTC)
Received: (qmail 45827 invoked by uid 500); 11 May 2014 02:56:01 -0000
Delivered-To: apmail-hadoop-mapreduce-user-archive@hadoop.apache.org
Received: (qmail 45704 invoked by uid 500); 11 May 2014 02:56:01 -0000
Mailing-List: contact user-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
List-Help: <mailto:user-help@hadoop.apache.org>
List-Unsubscribe: <mailto:user-unsubscribe@hadoop.apache.org>
List-Post: <mailto:user@hadoop.apache.org>
List-Id: <user.hadoop.apache.org>
Reply-To: user@hadoop.apache.org
Delivered-To: mailing list user@hadoop.apache.org
Received: (qmail 45453 invoked by uid 99); 11 May 2014 02:56:01 -0000
Received: from Unknown (HELO nike.apache.org) (192.87.106.230)
    by apache.org (qpsmtpd/0.29) with ESMTP; Sun, 11 May 2014 02:56:01 +0000
X-ASF-Spam-Status: No, hits=1.5 required=5.0
	tests=HTML_MESSAGE,RCVD_IN_DNSWL_LOW,SPF_PASS
X-Spam-Check-By: apache.org
Received-SPF: pass (nike.apache.org: domain of hujianan@gmail.com designates
 209.85.213.49 as permitted sender)
Received: from [209.85.213.49] (HELO mail-yh0-f49.google.com) (209.85.213.49)
    by apache.org (qpsmtpd/0.29) with ESMTP; Sun, 11 May 2014 02:55:57 +0000
Received: by mail-yh0-f49.google.com with SMTP id c41so5155019yho.22
        for <user@hadoop.apache.org>; Sat, 10 May 2014 19:55:33 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed;
        d=gmail.com; s=20120113;
        h=mime-version:date:message-id:subject:from:to:content-type;
        bh=JSperpPUn/KXIgrGApij8GY3rD2fg55EZIVCF533gb0=;
        b=FgQLk8gpUovX31AWL4cdh1bAQ6K+RJ4Pz65/GPj6dCjSmafZksp46vczEWjN778Nh7
         yINQUnxkkCzHNXQ1DJ7GbnkxVTjaUAMHwXd1baqiFUHchTUeGM40nWxnH3Kept32sx2K
         UJQ0kXLMkT61KNNh8yHBIetACrwzT/ys8Iv13Lst8jhVRDpiDfASzvDWCsHeGds7tduh
         rpQ6Ct2JCFRZFjujhrFAxOlRYnmbfW5Sm7mBdFKZBTidrIdfphL/Jvos+XZbqEkVavYy
         iWGdYFO/KWzEPCR9mOMqU8j+O/PM03oAexrHewbpviAgboMmRvi5rpQrMZ7AuWPdzXv+
         E+1A==
MIME-Version: 1.0
X-Received: by 10.236.14.2 with SMTP id c2mr28399837yhc.73.1399776933569; Sat,
 10 May 2014 19:55:33 -0700 (PDT)
Received: by 10.170.138.67 with HTTP; Sat, 10 May 2014 19:55:33 -0700 (PDT)
Date: Sun, 11 May 2014 10:55:33 +0800
Message-ID: 
 <CAHD=qRTY3uyPh75epYY5y+ViN8x328NG+Hr7VUCOQmRjsMSsag@mail.gmail.com>
Subject: Rack awareness and pipeline write
From: jianan hu <hujianan@gmail.com>
To: user@hadoop.apache.org
Content-Type: multipart/alternative; boundary=089e0139fc84e3e90504f916f6ab
X-Virus-Checked: Checked by ClamAV on apache.org

--089e0139fc84e3e90504f916f6ab
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

Hi everyone,

See HDFS documents, It says "For the common case, when the replication
factor is three, HDFS=E2=80=99s placement policy is to put one replica on o=
ne node
in the local rack, another on a node in a different (remote) rack, and the
last on a different node in the same remote rack."

Assume there are two racks A and B. According to rack awareness, the first
block is put in rack A, and the the other two replicated blocks will be
pushed into rack B.

However, why not store the first and second replicas in the local rack (A),
and the last in a different remote rack (B)? Both two scenarios have same
network traffic. What's the disadvantage of it?

Thanks.

Best Regards,
Jianan

--089e0139fc84e3e90504f916f6ab
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><span style=3D"font-family:arial,sans-serif;font-size:14px=
">Hi everyone,</span><div style=3D"font-family:arial,sans-serif;font-size:1=
4px"><br></div><div style=3D"font-family:arial,sans-serif;font-size:14px">S=
ee HDFS documents, It says &quot;For the common case, when the replication =
factor is three, HDFS=E2=80=99s placement policy is to put one replica on o=
ne node in the local rack, another on a node in a different (remote) rack, =
and the last on a different node in the same remote rack.&quot;</div>
<div style=3D"font-family:arial,sans-serif;font-size:14px"><br></div><div s=
tyle=3D"font-family:arial,sans-serif;font-size:14px">Assume there are two r=
acks A and B. According to rack awareness, the first block is put in rack A=
, and the the other two replicated blocks will be pushed into rack B.</div>
<div style=3D"font-family:arial,sans-serif;font-size:14px"><br></div><div s=
tyle=3D"font-family:arial,sans-serif;font-size:14px">However, why not store=
 the first and second replicas in the local rack (A), and the last in a dif=
ferent remote rack (B)? Both two scenarios have same network traffic. What&=
#39;s the disadvantage of it?</div>
<div style=3D"font-family:arial,sans-serif;font-size:14px"><br></div><div s=
tyle=3D"font-family:arial,sans-serif;font-size:14px">Thanks.</div><div styl=
e=3D"font-family:arial,sans-serif;font-size:14px"><br></div><div style=3D"f=
ont-family:arial,sans-serif;font-size:14px">
Best Regards,</div><div style=3D"font-family:arial,sans-serif;font-size:14p=
x">Jianan</div></div>

--089e0139fc84e3e90504f916f6ab--