Received: from localhost (maia-3.hub.org [200.46.204.184]) by postgresql.org (Postfix) with ESMTP id 9BFE49FB2ED for ; Mon, 25 Jun 2007 08:47:42 -0300 (ADT) Received: from postgresql.org ([200.46.204.71]) by localhost (mx1.hub.org [200.46.204.184]) (amavisd-maia, port 10024) with ESMTP id 43696-02 for ; Mon, 25 Jun 2007 08:47:30 -0300 (ADT) X-Greylist: from auto-whitelisted by SQLgrey-1.7.5 Received: from floppy.pyrenet.fr (news.pyrenet.fr [194.116.145.2]) by postgresql.org (Postfix) with ESMTP id 0BBDC9FB2A3 for ; Mon, 25 Jun 2007 08:47:37 -0300 (ADT) Received: by floppy.pyrenet.fr (Postfix, from userid 106) id 4F84E34030; Mon, 25 Jun 2007 13:47:35 +0200 (MET DST) From: "Sabin Coanda" X-Newsgroups: pgsql.performance Subject: Re: PostgreSQL Configuration Tool for Dummies - feedback adjustable control Date: Mon, 25 Jun 2007 14:47:43 +0300 Organization: Hub.Org Networking Services Lines: 39 Message-ID: References: X-Complaints-To: usenet@news.hub.org X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 6.00.2900.2180 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.2180 X-RFC2646: Format=Flowed; Response To: pgsql-performance@postgresql.org X-Virus-Scanned: Maia Mailguard 1.0.1 X-Spam-Status: No, hits=0.046 tagged_above=0 required=5 tests=AWL=0.045, BAYES_50=0.001 X-Spam-Level: X-Archive-Number: 200706/566 X-Sequence-Number: 25599 wrote in message news:Pine.LNX.4.64.0706221144290.14942@asgard.lang.hm... > On Fri, 22 Jun 2007, Sabin Coanda wrote: > >> Instead of (or in addition to) configure dozens of settings, what do you >> say about a feedback adjustable control based on the existing system >> statistics and parsing logs (e.g >> http://pgfouine.projects.postgresql.org/index.html ) ? > > something like this would be useful for advanced tuneing, but the biggest > problem is that it's so difficult to fingoure out a starting point. bad > choices at the starting point can cause several orders of magnatude > difference in the database performsnce. In addition we know that the > current defaults are bad for just about everyone (we just can't decide > what better defaults would be) > You are right. But an automatic tool beeing able to take decisions by different inputs, would be able to set a startup configuration too, based on the hw/sw environment, and interactive user requirements. > this horrible starting point gives people a bad first impression that a > simple tool like what's being discussed can go a long way towards solving. > Well, I think to an automatic tool, not an utopian application good for everything. For instance the existing automatic daemon have some abilities, bat not all of the VACUUM command. I'm realistic that good things may be done in steps, not once. I would be super happy if an available automatic configuration tool would be able to set for the beginning just the shared_buffers or max_fsm_pages based on the available memory. Adjustments can be done later. Regards, Sabin