X-Original-To: pgsql-performance-postgresql.org@localhost.postgresql.org Received: from localhost (av.hub.org [200.46.204.144]) by svr1.postgresql.org (Postfix) with ESMTP id EED3FDB985 for ; Thu, 17 Nov 2005 16:58:48 -0400 (AST) Received: from svr1.postgresql.org ([200.46.204.71]) by localhost (av.hub.org [200.46.204.144]) (amavisd-new, port 10024) with ESMTP id 19485-08 for ; Thu, 17 Nov 2005 20:58:49 +0000 (GMT) X-Greylist: domain auto-whitelisted by SQLgrey- Received: from zproxy.gmail.com (zproxy.gmail.com [64.233.162.193]) by svr1.postgresql.org (Postfix) with ESMTP id 09A5EDB998 for ; Thu, 17 Nov 2005 16:58:45 -0400 (AST) Received: by zproxy.gmail.com with SMTP id x3so22639nzd for ; Thu, 17 Nov 2005 12:58:47 -0800 (PST) DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=beta; d=gmail.com; h=received:message-id:date:from:to:subject:in-reply-to:mime-version:content-type:references; b=bMk6fhrIIKdMvOq34pTVXdX6rb8TLGWCsNFXUDG81H9Kzyl8VQStuFGyIO62+XsRDdD0dlGNRu2YvGrKYuAqKvi4+vbPzIOAJVawbIT+jGS0PP3C/mReFPHLXcYFQLLfQ6W1CctTfbkDxb0EXdvFavm+UyDe7lstEUcktd5Tsis= Received: by 10.64.131.4 with SMTP id e4mr7798847qbd; Thu, 17 Nov 2005 12:58:46 -0800 (PST) Received: by 10.64.243.11 with HTTP; Thu, 17 Nov 2005 12:58:46 -0800 (PST) Message-ID: <38242de90511171258v22f334f3ua474a8aa99bdc29d@mail.gmail.com> Date: Thu, 17 Nov 2005 13:58:46 -0700 From: Joshua Marsh To: pgsql-performance@postgresql.org Subject: Re: Hardware/OS recommendations for large databases ( In-Reply-To: MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_13115_5972285.1132261126498" References: <3E37B936B592014B978C4415F90D662D01995F7B@MI8NYCMAIL06.Mi8.com> <33c6269f0511152115v1a34ab81j30f1f5df5a6c6956@mail.gmail.com> <33c6269f0511171148q7ea67482xd595ea5717956aa2@mail.gmail.com> X-Virus-Scanned: by amavisd-new at hub.org X-Spam-Status: No, score=0.001 required=5 tests=[HTML_MESSAGE=0.001] X-Spam-Score: 0.001 X-Spam-Level: X-Archive-Number: 200511/332 X-Sequence-Number: 15589 ------=_Part_13115_5972285.1132261126498 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Content-Disposition: inline On 11/17/05, William Yu wrote: > > > No argument there. But it's pointless if you are IO bound. > > Why would you just accept "we're IO bound, nothing we can do"? I'd do > everything in my power to make my app go from IO bound to CPU bound -- > whether by optimizing my code or buying more hardware. I can tell you if > our OLTP servers were IO bound, it would run like crap. Instead of < 1 > sec, we'd be looking at 5-10 seconds per "user transaction" and our > users would be screaming bloody murder. > > In theory, you can always convert your IO bound DB to CPU bound by > stuffing more and more RAM into your server. (Or partitioning the DB > across multiple servers.) Whether it's cost effective depends on the DB > and how much your users are paying you -- and that's a case-by-case > analysis. Not a global statement of "IO-bound, pointless". We all want our systems to be CPU bound, but it's not always possible. Remember, he is managing a 5 TB Databse. That's quite a bit different than = a 100 GB or even 500 GB database. ------=_Part_13115_5972285.1132261126498 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Content-Disposition: inline
On 11/17/05, William Yu <wyu@talisys.com<= /a>> wrote:
> No argument there.  But it's pointless if you are IO bound.<= br>
Why would you just accept "we're IO bound, nothing we can do&qu= ot;? I'd do
everything in my power to make my app go from IO bound to CP= U bound --
whether by optimizing my code or buying more hardware. I can tell you i= f
our OLTP servers were IO bound, it would run like crap. Instead of <= ; 1
sec, we'd be looking at 5-10 seconds per "user transaction"= ; and our
users would be screaming bloody murder.

In theory, you can alway= s convert your IO bound DB to CPU bound by
stuffing more and more RAM in= to your server. (Or partitioning the DB
across multiple servers.) Whethe= r it's cost effective depends on the DB
and how much your users are paying you -- and that's a case-by-case
= analysis. Not a global statement of "IO-bound, pointless".

We all want our systems to be CPU bound, but it's not always possible.  Remember, he is managing a 5 TB Databse.  That's quite a bit different than a 100 GB or even 500 GB database.
 

------=_Part_13115_5972285.1132261126498--