X-Original-To: pgsql-performance-postgresql.org@localhost.postgresql.org Received: from localhost (unknown [200.46.204.2]) by svr1.postgresql.org (Postfix) with ESMTP id 9F64FD1B4BE for ; Tue, 6 Apr 2004 02:46:45 -0300 (ADT) Received: from svr1.postgresql.org ([200.46.204.71]) by localhost (neptune.hub.org [200.46.204.2]) (amavisd-new, port 10024) with ESMTP id 53968-08 for ; Tue, 6 Apr 2004 02:47:24 -0300 (ADT) Received: from sss.pgh.pa.us (sss.pgh.pa.us [66.207.139.130]) by svr1.postgresql.org (Postfix) with ESMTP id E9B03D1B4B6 for ; Tue, 6 Apr 2004 02:46:38 -0300 (ADT) Received: from sss2.sss.pgh.pa.us (tgl@localhost [127.0.0.1]) by sss.pgh.pa.us (8.12.11/8.12.11) with ESMTP id i365lNvI019782; Tue, 6 Apr 2004 01:47:23 -0400 (EDT) To: Qing Zhao Cc: pgsql-performance@postgresql.org Subject: Re: possible improvement between G4 and G5 In-reply-to: References: <009901c41b48$8357c860$2500fa0a@CardServices.TCI.com> Comments: In-reply-to Qing Zhao message dated "Mon, 05 Apr 2004 17:53:49 -0700" Date: Tue, 06 Apr 2004 01:47:22 -0400 Message-ID: <19781.1081230442@sss.pgh.pa.us> From: Tom Lane X-Virus-Scanned: by amavisd-new at postgresql.org X-Archive-Number: 200404/55 X-Sequence-Number: 6355 Qing Zhao writes: > We have got a G5 64-bit processor to replace an old G4 32-bit > processor. Given everything else equal, should we see a big > improvement on PG's performance? Nope. Database performance typically depends on disk performance first, and RAM size second. A 64-bit processor might help by allowing you to install more RAM, but you didn't say that you had. > The other question I have is that, when I tried different size for > shared_buffer ( i used 10,000, 1,000, 528, 256) and Max > connections=32, it gives me error when I tried to start PG using > pg_ctl start as postgres. It kept saying this is bigger than the > system Shared Memory. Out-of-the-box, Mac OS X has a very low SHMMAX limit. See the PG admin docs or the mail list archives about how to increase it. You should do this --- most people find that you want to set shared_buffers to 1000 or 10000 or so for best performance. regards, tom lane