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 8D152DBAD3 for ; Thu, 17 Nov 2005 17:23:00 -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 41187-03 for ; Thu, 17 Nov 2005 21:23:00 +0000 (GMT) X-Greylist: from auto-whitelisted by SQLgrey- Received: from floppy.pyrenet.fr (news.pyrenet.fr [194.116.145.2]) by svr1.postgresql.org (Postfix) with ESMTP id 5F56BDBA78 for ; Thu, 17 Nov 2005 17:22:54 -0400 (AST) Received: by floppy.pyrenet.fr (Postfix, from userid 106) id 10F0331059; Thu, 17 Nov 2005 22:22:56 +0100 (MET) From: William Yu X-Newsgroups: pgsql.performance Subject: Re: Hardware/OS recommendations for large databases ( Date: Thu, 17 Nov 2005 13:22:47 -0800 Organization: Hub.Org Networking Services Lines: 32 Message-ID: References: <3E37B936B592014B978C4415F90D662D01995F7B@MI8NYCMAIL06.Mi8.com> <33c6269f0511152115v1a34ab81j30f1f5df5a6c6956@mail.gmail.com> <33c6269f0511171148q7ea67482xd595ea5717956aa2@mail.gmail.com> <38242de90511171258v22f334f3ua474a8aa99bdc29d@mail.gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Complaints-To: usenet@news.hub.org User-Agent: Mozilla Thunderbird 1.0 (X11/20041206) X-Accept-Language: en-us, en In-Reply-To: <38242de90511171258v22f334f3ua474a8aa99bdc29d@mail.gmail.com> To: pgsql-performance@postgresql.org X-Virus-Scanned: by amavisd-new at hub.org X-Spam-Status: No, score=0 required=5 tests=[AWL=0.000] X-Spam-Score: 0 X-Spam-Level: X-Archive-Number: 200511/335 X-Sequence-Number: 15592 Joshua Marsh wrote: > > 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. I did say "in theory". :) I'm pretty sure google is more CPU bound than IO bound -- they just spread their DB over 50K servers or whatever. Not everybody is willing to pay for that but it's always in the realm of plausibility. Plus we have to go back to the statement I was replying to which was "I have yet to come across a DB system that wasn't IO bound".