Received: from localhost (unknown [200.46.208.211]) by mail.postgresql.org (Postfix) with ESMTP id 01D006325DE for ; Thu, 14 Jan 2010 16:03:04 -0400 (AST) Received: from mail.postgresql.org ([200.46.204.86]) by localhost (mx1.hub.org [200.46.208.211]) (amavisd-maia, port 10024) with ESMTP id 62736-04 for ; Thu, 14 Jan 2010 20:02:49 +0000 (UTC) X-Greylist: from auto-whitelisted by SQLgrey-1.7.6 Received: from pyxis.powervt.com.ar (quportal.powervt.com.ar [190.11.152.66]) by mail.postgresql.org (Postfix) with ESMTP id 93CD4632771 for ; Thu, 14 Jan 2010 16:02:52 -0400 (AST) Received: from pcfhevia ([186.124.123.2]) by pyxis.powervt.com.ar (IPGroup Mail Server) with ASMTP id 1OT48953 for ; Thu, 14 Jan 2010 17:02:44 -0300 From: "Fernando Hevia" To: Subject: new server I/O setup Date: Thu, 14 Jan 2010 17:03:52 -0300 Organization: IP-TEL Message-ID: <3CD96C0AB4114484BFE358A3EDE61C01@iptel.com.ar> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_0061_01CA953B.8C4EE430" X-Mailer: Microsoft Office Outlook 11 Thread-Index: AcqVVLFsurWpk+jQRvutD4rsy6b81A== X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.5579 X-Virus-Scanned: Maia Mailguard 1.0.1 X-Spam-Status: No, hits=0.23 tagged_above=-10 required=5 tests=AWL=0.228, BAYES_50=0.001, HTML_MESSAGE=0.001 X-Spam-Level: X-Archive-Number: 201001/304 X-Sequence-Number: 37184 This is a multi-part message in MIME format. ------=_NextPart_000_0061_01CA953B.8C4EE430 Content-Type: text/plain; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable Hi all, =20 I've just received this new server: 1 x XEON 5520 Quad Core w/ HT 8 GB RAM 1066 MHz 16 x SATA II Seagate Barracuda 7200.12 3ware 9650SE w/ 256MB BBU =20 It will run an Ubuntu 8.04 LTS Postgres 8.4 dedicated server. Its = database will be getting between 100 and 1000 inserts per second (those are call detail records of ~300 bytes each) of around 20 clients (voip gateways). Other activity is mostly read-only and some non time-critical writes generally at off peak hours. =20 So my first choice was: =20 2 discs in RAID 1 for OS + pg_xlog partitioned with ext2. 12 discs in RAID 10 for postgres data, sole partition with ext3. 2 spares =20 =20 My second choice is: =20 4 discs in RAID 10 for OS + pg_xlog partitioned with ext2 10 discs in RAID 10 for postgres, ext3 2 spares. =20 The bbu cach=E9 will be enabled for both raid volumes. =20 I justified my first choice in that WAL writes are sequentially and OS pretty much are too, so a RAID 1 probably would hold ground against a 12 disc RAID 10 with random writes. =20 I don't know in advance if I will manage to gather enough time to try = out both setups so I wanted to know what you guys think of these 2 = alternatives. Do you think a single RAID 1 will become a bottleneck? Feel free to = suggest a better setup I hadn't considered, it would be most welcome. =20 Pd: any clue if hdparm works to deactive the disks write cache even if = they are behind the 3ware controller? =20 Regards, Fernando. =20 ------=_NextPart_000_0061_01CA953B.8C4EE430 Content-Type: text/html; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable

Hi=20 all,

I've = just=20 received this new server:

1 x = XEON 5520 Quad=20 Core w/ HT

8 GB = RAM 1066=20 MHz

16 x = SATA II Seagate=20 Barracuda 7200.12

3ware = 9650SE w/=20 256MB BBU

It = will run an=20 Ubuntu 8.04 LTS Postgres 8.4 dedicated server. = Its database will=20 be getting between 100 and 1000 inserts per second (those are call = detail=20 records of ~300 bytes each) of around 20 clients (voip gateways). Other = activity=20 is mostly read-only and some non time-critical writes generally at = off peak=20 hours.

So=20 my=20 first choice was:

2 discs in RAID=20 1 for OS + pg_xlog partitioned with ext2.

12 = discs in RAID 10=20 for postgres data, sole partition with ext3.

2=20 spares

My = second choice=20 is:

4 = discs in RAID 10=20 for OS + pg_xlog partitioned with ext2

10 = discs in RAID 10=20 for postgres, ext3

2=20 spares.

The = bbu cach=E9 will=20 be enabled for both raid volumes.

I = justified my first=20 choice in that WAL writes are sequentially and OS pretty much are = too, so a=20 RAID 1 probably would hold ground against a 12 disc RAID 10 with = random=20 writes.

I = don't know in=20 advance if I will manage to gather enough time to try out both = setups so I=20 wanted to know what you guys think of these 2 alternatives. Do you = think a=20 single RAID 1 will become a bottleneck? Feel free to suggest a = better setup=20 I hadn't considered, it would be most welcome.

Pd: = any clue if=20 hdparm works to deactive the disks write cache even if they are = behind the=20 3ware controller?

Regards,

Fernando.

------=_NextPart_000_0061_01CA953B.8C4EE430--