Return-Path: 
 <user-return-10435-apmail-hadoop-hdfs-user-archive=hadoop.apache.org@hadoop.apache.org>
X-Original-To: apmail-hadoop-hdfs-user-archive@minotaur.apache.org
Delivered-To: apmail-hadoop-hdfs-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 DA57B107CB
	for <apmail-hadoop-hdfs-user-archive@minotaur.apache.org>;
 Thu, 29 Aug 2013 21:34:28 +0000 (UTC)
Received: (qmail 27853 invoked by uid 500); 29 Aug 2013 21:34:23 -0000
Delivered-To: apmail-hadoop-hdfs-user-archive@hadoop.apache.org
Received: (qmail 27566 invoked by uid 500); 29 Aug 2013 21:34:23 -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 27559 invoked by uid 99); 29 Aug 2013 21:34:23 -0000
Received: from athena.apache.org (HELO athena.apache.org) (140.211.11.136)
    by apache.org (qpsmtpd/0.29) with ESMTP; Thu, 29 Aug 2013 21:34:23 +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 (athena.apache.org: domain of secsubs@gmail.com designates
 209.85.212.68 as permitted sender)
Received: from [209.85.212.68] (HELO mail-vb0-f68.google.com) (209.85.212.68)
    by apache.org (qpsmtpd/0.29) with ESMTP; Thu, 29 Aug 2013 21:34:18 +0000
Received: by mail-vb0-f68.google.com with SMTP id e13so132519vbg.7
        for <user@hadoop.apache.org>; Thu, 29 Aug 2013 14:33:57 -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=jfWP/7YMekM8admtdhlCdMOQ5UzIC3M/qz140ExTZ2g=;
        b=a/hV25DG9l11SK9699ZhHyG5oimJAjEURVl2sUUMtI1+4tdZWkJHVwzWocWuc4w2LD
         5K7oQ2RvjmYpwmTCs+g+tQ9qdbOzvTYUTSeCg/b+moxFeoogd54IyDiMyFESLV1xpQIi
         eUR00FSxfS8knCLoGabAUkCRTVgUlPHUq/rlqbh5CASjNUMwkUMVw7SSZBKd3L1BrpkY
         PED3IEqFeiC7iduZcjY5RzHdeynt2cmX+u8sxw74BlUWRHErtiH0LleFxTyq+JQ8mOka
         WImSFP1m0W8fnklkwgUVaGJXMSSv8yl1NT0XnxuBn3PFXGKuTfFtbJUMBpzEDbWLiXL1
         AEXA==
MIME-Version: 1.0
X-Received: by 10.220.145.132 with SMTP id d4mr4932468vcv.9.1377812037860;
 Thu, 29 Aug 2013 14:33:57 -0700 (PDT)
Received: by 10.221.57.129 with HTTP; Thu, 29 Aug 2013 14:33:57 -0700 (PDT)
Date: Thu, 29 Aug 2013 14:33:57 -0700
Message-ID: 
 <CADPi3fi09M7aqKrsfc07sdu8QOXgZS+pFb-RDQ7cLf8pa4nOgw@mail.gmail.com>
Subject: TB per core sweet spot
From: Xuri Nagarin <secsubs@gmail.com>
To: user@hadoop.apache.org
Content-Type: multipart/alternative; boundary=047d7b34347c15832004e51cdd5e
X-Virus-Checked: Checked by ClamAV on apache.org

--047d7b34347c15832004e51cdd5e
Content-Type: text/plain; charset=ISO-8859-1

Hi,

I realize there is no perfect spec for data nodes as lot depends on use
cases and work loads but I am curious if there are any rules of thumb or
no-go zones in terms of how many terabytes per core is ok?

So a few questions assuming 1 core per hdd holds:
Is there a no-go zone in terms of tb/core? I ask because I am seeing
4TB/core nodes in some of the clusters and wondering if that's too much?
Does tb/core depend on the core speed? For example, while a 1.8Ghz might be
able to handle 1TB, going to 4TB requires a 3.6Ghz E5 Xeon core?
Dramatic difference between Xeon E3 vs E5 or incremental?
Any comments on disk choice - SATA vs SAS, 5.9k vs 7.2k vs 10k, SATA2 vs 3?

Again, I realize there is a huge YMMV factor here but I would love to hear
experiences or research people have done before picking specs for their
nodes including vendors/models.


Thanks,

Xuri

--047d7b34347c15832004e51cdd5e
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div><div><div><div><div><div><div><div><div>Hi,<br><br></=
div>I realize there is no perfect spec for data nodes as lot depends on use=
 cases and work loads but I am curious if there are any rules of thumb or n=
o-go zones in terms of how many terabytes per core is ok?<br>
<br></div>So a few questions assuming 1 core per hdd holds:<br></div>Is the=
re a no-go zone in terms of tb/core? I ask because I am seeing 4TB/core nod=
es in some of the clusters and wondering if that&#39;s too much?<br></div>
Does tb/core depend on the core speed? For example, while a 1.8Ghz might be=
 able to handle 1TB, going to 4TB requires a 3.6Ghz E5 Xeon core?<br></div>=
Dramatic difference between Xeon E3 vs E5 or incremental?<br></div>Any comm=
ents on disk choice - SATA vs SAS, 5.9k vs 7.2k vs 10k, SATA2 vs 3?<br>
<br></div>Again, I realize there is a huge YMMV factor here but I would lov=
e to hear experiences or research people have done before picking specs for=
 their nodes including vendors/models.<br><br><br></div>Thanks,<br><br>
</div>Xuri<br><br><br></div>

--047d7b34347c15832004e51cdd5e--