X-Original-To: pgsql-admin-postgresql.org@localhost.postgresql.org Received: from localhost (unknown [200.46.204.144]) by svr1.postgresql.org (Postfix) with ESMTP id 04DE0534AD for ; Wed, 30 Mar 2005 00:55:38 +0100 (BST) 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 35835-04 for ; Tue, 29 Mar 2005 23:55:37 +0000 (GMT) Received: from koolancexeon.g2switchworks.com (mail.g2switchworks.com [63.87.162.25]) by svr1.postgresql.org (Postfix) with ESMTP id B349953448 for ; Wed, 30 Mar 2005 00:55:35 +0100 (BST) Received: mail.g2switchworks.com 10.10.1.8 from 10.10.1.37 10.10.1.37 via HTTP with MS-WebStorage 6.5.6944 Received: from state.g2switchworks.com by mail.g2switchworks.com; 29 Mar 2005 17:55:38 -0600 Subject: Re: Replication solutions high volume db From: Scott Marlowe To: David B Cc: pgsql-admin@postgresql.org In-Reply-To: References: Content-Type: text/plain Content-Transfer-Encoding: 7bit Message-Id: <1112140537.31277.15.camel@state.g2switchworks.com> Mime-Version: 1.0 X-Mailer: Ximian Evolution 1.4.6 (1.4.6-2) Date: Tue, 29 Mar 2005 17:55:38 -0600 X-Virus-Scanned: by amavisd-new at hub.org X-Spam-Status: No, hits=0.004 tagged_above=0 required=5 tests=AWL X-Spam-Level: X-Archive-Number: 200503/470 X-Sequence-Number: 16968 On Tue, 2005-03-29 at 17:34, David B wrote: > Folks, > > I'm thinking of replication tools for high volume site. > Postgres on dual Opteron box. 1/2 GB of data. > > Big overnight load of data. > Some inserts during the data but mostly selects. > > I think I've narrowed it down to Slony and Mammoth. > Any pro's or con's on one vs the other? > > There is the obvious price difference but thats managable if Mammoth > substancially better. > > Some area where our environement might stretch a solution. > Some records can be large. Limitied to one table but they can be 40K+ > of chraracters in a column. > > High volume of inserts coming during nightly load (millions of rows). > These loads are coming in commited blocks of 100. > > That cannot slow down the replication. > Our call center agents needs access to them asap. > > Any locking issues on the master during the replication process? > > > Anything else I should be concerned about or things I need to consider? > Personal experiences you have had running either. I don't see any obvious problems for either replication engine here. Since slony is free and pretty easy to setup, I'd make a test setup and see how it handles the load. Generally speaking I'd expect Mammoth to handle high loads a little better since it's a log shipper and not trigger based. So, if slony proves too slow for this situation, then I'd recommend asking the folks at Command Prompt for a demo of their system to see if it can.