X-Original-To: pgsql-performance-postgresql.org@postgresql.org
Received: from localhost (mx1.hub.org [200.46.208.251])
	by postgresql.org (Postfix) with ESMTP id EA1299FB29E
	for <pgsql-performance-postgresql.org@postgresql.org>;
	Thu,  3 Aug 2006 02:36:48 -0300 (ADT)
Received: from postgresql.org ([200.46.204.71])
	by localhost (mx1.hub.org [200.46.208.251]) (amavisd-new, port 10024)
	with ESMTP id 34604-02 for
	<pgsql-performance-postgresql.org@postgresql.org>;
	Thu,  3 Aug 2006 02:36:44 -0300 (ADT)
X-Greylist: domain auto-whitelisted by SQLgrey-
Received: from nz-out-0102.google.com (nz-out-0102.google.com
 [64.233.162.197])
	by postgresql.org (Postfix) with ESMTP id 74AE69FB29A
	for <pgsql-performance@postgresql.org>;
	Thu,  3 Aug 2006 02:36:45 -0300 (ADT)
Received: by nz-out-0102.google.com with SMTP id z31so442208nzd
	for <pgsql-performance@postgresql.org>;
	Wed, 02 Aug 2006 22:36:40 -0700 (PDT)
DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=beta; d=gmail.com;
	h=received:message-id:date:from:sender:to:subject:cc:in-reply-to:mime-version:content-type:references:x-google-sender-auth;
	b=pOvZtf8l0E1NNl6KrbhFQwVGXyirRSVPPHCiPSsxN8BU7Tf/hysAidcN7fOFbaUoU6V/ywZws5sTGxlI2lLLpKz3iKv0c4qPsjzwAFFeb6ny2etyTgg2U0ej/NpUfr0DTeKR5mTdyH052q8W+MfNGZGA2a3kpMKyGhZu1F3gXWA=
Received: by 10.65.112.5 with SMTP id p5mr2571469qbm;
	Wed, 02 Aug 2006 22:36:40 -0700 (PDT)
Received: by 10.64.179.18 with HTTP; Wed, 2 Aug 2006 22:36:10 -0700 (PDT)
Message-ID: <5f820f750608022236yd8dd6escce3bb849b91bddb@mail.gmail.com>
Date: Thu, 3 Aug 2006 01:36:10 -0400
From: "Denis Lussier" <denisl@enterprisedb.com>
To: "Luke Lonergan" <llonergan@greenplum.com>
Subject: Re: XFS filessystem for Datawarehousing -2
Cc: "Milen Kulev" <makulev@gmx.net>, pgsql-performance@postgresql.org
In-Reply-To: <C0F55F94.2B857%llonergan@greenplum.com>
MIME-Version: 1.0
Content-Type: multipart/alternative;
	boundary="----=_Part_67512_30951560.1154583370571"
References: <014401c6b5b8$98926a20$0a00a8c0@trivadis.com>
	<C0F55F94.2B857%llonergan@greenplum.com>
X-Google-Sender-Auth: 6507723b2c1d39ff
X-Virus-Scanned: Maia Mailguard 1.0.1
X-Spam-Status: No, hits=0.544 tagged_above=0 required=5 tests=AWL,
	DNS_FROM_RFC_ABUSE, HTML_30_40, HTML_MESSAGE, SPF_PASS
X-Spam-Level: 
X-Archive-Number: 200608/37
X-Sequence-Number: 20222

------=_Part_67512_30951560.1154583370571
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

I was kinda thinking that making the Block Size configurable at InitDB time
would be a nice & simple enhancement for PG 8.3.  My own personal rule of
thumb for sizing is 8k for OLTP, 16k for mixed use, & 32k for DWH.

I have no personal experience with XFS, but, I've seen numerous internal
edb-postgres test results that show that of all file systems... OCFS
2.0seems to be quite good for PG update intensive apps (especially on
64 bit
machines).

On 8/1/06, Luke Lonergan <llonergan@greenplum.com> wrote:
>
> Milen,
>
> On 8/1/06 3:19 PM, "Milen Kulev" <makulev@gmx.net> wrote:
>
> > Sorry, forgot to ask:
> > What is the recommended/best  PG block size for DWH  database?  16k,
> 32k, 64k
> > ?
> > What hsould be the relation  between XFS/RAID stripe size and PG block
> size ?
>
> We have found that the page size in PG starts to matter only at very high
> disk performance levels around 1000MB/s.  Other posters have talked about
> maintenance tasks improving in performance, but I haven't seen it.
>
> - Luke
>
>
>
> ---------------------------(end of broadcast)---------------------------
> TIP 4: Have you searched our list archives?
>
>                http://archives.postgresql.org
>

------=_Part_67512_30951560.1154583370571
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

<br>I was kinda thinking that making the Block Size configurable at InitDB time would be a nice &amp; simple enhancement for PG 8.3.&nbsp; My own personal rule of thumb for sizing is 8k for OLTP, 16k for mixed use, &amp; 32k for DWH.
<br><br>I have no personal experience with XFS, but, I've seen numerous internal edb-postgres test results that show that of all file systems... OCFS 2.0 seems to be quite good for PG update intensive apps (especially on 64 bit machines).
<br><br><div><span class="gmail_quote">On 8/1/06, <b class="gmail_sendername">Luke Lonergan</b> &lt;<a href="mailto:llonergan@greenplum.com">llonergan@greenplum.com</a>&gt; wrote:</span><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
Milen,<br><br>On 8/1/06 3:19 PM, &quot;Milen Kulev&quot; &lt;<a href="mailto:makulev@gmx.net">makulev@gmx.net</a>&gt; wrote:<br><br>&gt; Sorry, forgot to ask:<br>&gt; What is the recommended/best&nbsp;&nbsp;PG block size for DWH&nbsp;&nbsp;database?&nbsp;&nbsp;16k, 32k, 64k
<br>&gt; ?<br>&gt; What hsould be the relation&nbsp;&nbsp;between XFS/RAID stripe size and PG block size ?<br><br>We have found that the page size in PG starts to matter only at very high<br>disk performance levels around 1000MB/s.&nbsp;&nbsp;Other posters have talked about
<br>maintenance tasks improving in performance, but I haven't seen it.<br><br>- Luke<br><br><br><br>---------------------------(end of broadcast)---------------------------<br>TIP 4: Have you searched our list archives?<br>
<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; <a href="http://archives.postgresql.org">http://archives.postgresql.org</a><br></blockquote></div><br>

------=_Part_67512_30951560.1154583370571--