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 9F2BE52861 for ; Sat, 21 May 2005 12:16:07 -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 64919-10 for ; Sat, 21 May 2005 15:16:01 +0000 (GMT) Received: from smtp-out3.blueyonder.co.uk (smtp-out3.blueyonder.co.uk [195.188.213.6]) by svr1.postgresql.org (Postfix) with ESMTP id 4A3B752859 for ; Sat, 21 May 2005 12:15:59 -0300 (ADT) Received: from [192.168.1.10] ([82.43.93.192]) by smtp-out3.blueyonder.co.uk with Microsoft SMTPSVC(5.0.2195.6713); Sat, 21 May 2005 16:17:10 +0100 In-Reply-To: <428D4146.6000203@familyhealth.com.au> References: <428D1153.1020109@rueb.com> <428D4146.6000203@familyhealth.com.au> Mime-Version: 1.0 (Apple Message framework v730) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Message-Id: <803159ED-9B03-43E0-ACE6-DC40E73D55C1@advfn.com> Cc: Steve Bergman , pgsql-performance@postgresql.org Content-Transfer-Encoding: 7bit From: Alex Stapleton Subject: Re: Optimizing for writes. Data integrity not critical Date: Sat, 21 May 2005 16:16:30 +0100 To: Christopher Kings-Lynne X-Mailer: Apple Mail (2.730) X-OriginalArrivalTime: 21 May 2005 15:17:11.0015 (UTC) FILETIME=[2901FF70:01C55E18] X-Virus-Scanned: by amavisd-new at hub.org X-Spam-Status: No, hits=0 tagged_above=0 required=5 tests= X-Spam-Level: X-Archive-Number: 200505/340 X-Sequence-Number: 12504 Is using a ramdisk in situations like this entirely ill-advised then? When data integrity isn't a huge issue and you really need good write performance it seems like it wouldn't hurt too much. Unless I am missing something? On 20 May 2005, at 02:45, Christopher Kings-Lynne wrote: >> I'm doing the writes individually. Is there a better way? >> Combining them all into a transaction or something? >> > > Use COPY of course :) > > Or at worst bundle 1000 inserts at a time in a transation... > > And if you seriously do not care about your data at all, set fsync > = off in you postgresql.conf for a mega speedup. > > Chris > > ---------------------------(end of > broadcast)--------------------------- > TIP 1: subscribe and unsubscribe commands go to > majordomo@postgresql.org > >