Received: from magus.postgresql.org (magus.postgresql.org [87.238.57.229]) by mail.postgresql.org (Postfix) with ESMTP id CE750110908A for ; Wed, 8 Feb 2012 22:35:47 -0400 (AST) Received: from ns3.uci.cu ([200.55.140.180] helo=mx3.uci.cu) by magus.postgresql.org with smtp (Exim 4.72) (envelope-from ) id 1RvJrH-0001DJ-Av for pgsql-performance@postgresql.org; Thu, 09 Feb 2012 02:35:46 +0000 Received: (qmail 9277 invoked by uid 507); 9 Feb 2012 02:35:13 -0000 Received: from 10.0.0.188 by ns3.uci.cu (envelope-from , uid 501) with qmail-scanner-2.01st (avp: 5.0.2.0. spamassassin: 3.0.6. perlscan: 2.01st. Clear:RC:1(10.0.0.188):. Processed in 0.973964 secs); 09 Feb 2012 02:35:13 -0000 Received: from unknown (HELO ucimail5.uci.cu) (10.0.0.188) by 0 with SMTP; 9 Feb 2012 02:35:12 -0000 Received: from localhost (localhost.localdomain [127.0.0.1]) by ucimail5.uci.cu (Postfix) with ESMTP id 55A9E1103EC for ; Wed, 8 Feb 2012 21:35:12 -0500 (CST) X-Virus-Scanned: amavisd-new at uci.cu Received: from ucimail5.uci.cu ([127.0.0.1]) by localhost (ucimail5.uci.cu [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id R74pgSyXDpAI for ; Wed, 8 Feb 2012 21:35:10 -0500 (CST) Received: by ucimail5.uci.cu (Postfix, from userid 101) id A8EB322C002; Wed, 8 Feb 2012 21:35:10 -0500 (CST) Received: from [10.8.26.2] (unknown [10.8.26.2]) (Authenticated sender: mlortiz@uci.cu) by ucimail5.uci.cu (Postfix) with ESMTPSA id 2F2D11103E1; Wed, 8 Feb 2012 21:35:10 -0500 (CST) Message-ID: <4F3329DF.2060805@uci.cu> Date: Wed, 08 Feb 2012 21:35:19 -0430 From: Marcos Ortiz Valmaseda User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:8.0) Gecko/20111115 Thunderbird/8.0 MIME-Version: 1.0 To: Peter van Hardenberg CC: Scott Marlowe , Josh Berkus , pgsql-performance@postgresql.org Subject: Re: random_page_cost = 2.0 on Heroku Postgres References: <4F331866.7070401@agliodbs.com> In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 8bit X-Pg-Spam-Score: -1.9 (-) X-Archive-Number: 201202/67 X-Sequence-Number: 46239 On 08/02/12 21:15, Peter van Hardenberg wrote: > Having read the thread, I don't really see how I could study what a > more principled value would be. > > That said, I have access to a very large fleet in which to can collect > data so I'm all ears for suggestions about how to measure and would > gladly share the results with the list. > > Peter > > On Wed, Feb 8, 2012 at 5:39 PM, Scott Marlowe wrote: >> On Wed, Feb 8, 2012 at 5:50 PM, Josh Berkus wrote: >>> On 2/7/12 4:59 PM, Peter van Hardenberg wrote: >>>> Per the thread from last month, I've updated the default >>>> random_page_cost on Heroku Postgres to reduce the expected cost of a >>>> random_page on all new databases. >>> This is because Heroku uses AWS storage, which has fast seeks but poor >>> throughput compared to internal disk on a standard system, BTW. >> Also judging by the other thread, it might be something to stop closer >> to 1.2 to 1.4 or something. >> >> -- >> Sent via pgsql-performance mailing list (pgsql-performance@postgresql.org) >> To make changes to your subscription: >> http://www.postgresql.org/mailpref/pgsql-performance > > You can execute several queries with the three different values provided by Scott and Josh. - SET random_page_cost = 2.0 First execution of the queries with EXPLAIN ANALYZE - SET random_page_cost = 1.4 Second execution of the queries with EXPLAIN ANALYZE - SET random_page_cost = 1.2 Second execution of the queries with EXPLAIN ANALYZE And then, you can compare the pattern behind these queries executions Regards, -- Marcos Luis Ortíz Valmaseda Sr. Software Engineer (UCI) http://marcosluis2186.posterous.com http://www.linkedin.com/in/marcosluis2186 Twitter: @marcosluis2186 Fin a la injusticia, LIBERTAD AHORA A NUESTROS CINCO COMPATRIOTAS QUE SE ENCUENTRAN INJUSTAMENTE EN PRISIONES DE LOS EEUU! http://www.antiterroristas.cu http://justiciaparaloscinco.wordpress.com