X-Original-To: pgsql-performance-postgresql.org@localhost.postgresql.org Received: from localhost (unknown [200.46.204.144]) by svr1.postgresql.org (Postfix) with ESMTP id 2A3B6542C9 for ; Tue, 10 May 2005 11:16:32 -0300 (ADT) 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 49047-02 for ; Tue, 10 May 2005 14:16:20 +0000 (GMT) Received: from sss.pgh.pa.us (sss.pgh.pa.us [66.207.139.130]) by svr1.postgresql.org (Postfix) with ESMTP id 9DF60542E8 for ; Tue, 10 May 2005 11:16:17 -0300 (ADT) Received: from sss2.sss.pgh.pa.us (tgl@localhost [127.0.0.1]) by sss.pgh.pa.us (8.13.1/8.13.1) with ESMTP id j4AEGKgb010062; Tue, 10 May 2005 10:16:20 -0400 (EDT) To: molson@oceanconsulting.com Cc: pgsql-performance@postgresql.org Subject: Re: Prefetch In-reply-to: <200505100652.51401.molson@oceanconsulting.com> References: <200505100652.51401.molson@oceanconsulting.com> Comments: In-reply-to Matt Olson message dated "Tue, 10 May 2005 06:52:51 -0700" Date: Tue, 10 May 2005 10:16:20 -0400 Message-ID: <10061.1115734580@sss.pgh.pa.us> From: Tom Lane X-Virus-Scanned: by amavisd-new at hub.org X-Spam-Status: No, hits=0.007 tagged_above=0 required=5 tests=AWL X-Spam-Level: X-Archive-Number: 200505/138 X-Sequence-Number: 12302 Matt Olson writes: > Other databases like Oracle and DB2 implement some sort of row prefetch. Has > there been serious consideration of implementing something like a prefetch > subsystem? No. > Does anyone have any opinions as to why this would be a bad idea for > postgres? We know even less than the OS does about disk layout, and not a lot more than it about what our next request will be. (If we're doing a seqscan, then of course that's not true, but I would expect the OS to be able to figure that one out and do readahead.) You haven't shown us your problem queries, but I think that conventional query tuning would be a more appropriate answer. In particular I wonder whether you shouldn't be looking at ways to calculate multiple aggregates in parallel. regards, tom lane