X-Original-To: pgsql-performance-postgresql.org@localhost.postgresql.org Received: from localhost (unknown [64.117.224.130]) by svr1.postgresql.org (Postfix) with ESMTP id C5A48D1B4E9 for ; Wed, 8 Oct 2003 19:08:23 +0000 (GMT) Received: from svr1.postgresql.org ([64.117.224.193]) by localhost (neptune.hub.org [64.117.224.130]) (amavisd-new, port 10024) with ESMTP id 21169-09 for ; Wed, 8 Oct 2003 16:07:39 -0300 (ADT) Received: from colsweeper.cranel.com (newmail.cranel.com [12.32.71.147]) by svr1.postgresql.org (Postfix) with ESMTP id A856BD1B4F0 for ; Wed, 8 Oct 2003 16:07:37 -0300 (ADT) Received: from colmail01.cranel.com (colmail01.cranel.com) by colsweeper.cranel.com (Content Technologies SMTPRS 4.2.5) with ESMTP id ; Wed, 8 Oct 2003 15:09:43 -0400 Received: from cranel.com (gspiegelberg.cranel.com [192.168.11.134]) by colmail01.cranel.com with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2653.13) id T6NHFH6G; Wed, 8 Oct 2003 15:10:26 -0400 Message-ID: <3F846072.3090700@cranel.com> Date: Wed, 08 Oct 2003 15:07:30 -0400 From: Greg Spiegelberg Organization: Cranel, Inc. User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.4) Gecko/20030624 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Dror Matalon Cc: PgSQL Performance ML Subject: Re: Compare rows References: <3F841A28.8060504@cranel.com> <200310080901.45093.josh@agliodbs.com> <3F843AFD.3020700@cranel.com> <3F843F73.6030302@joeconway.com> <3F8459FA.1030008@cranel.com> <20031008185527.GK2979@rlx11.zapatec.com> In-Reply-To: <20031008185527.GK2979@rlx11.zapatec.com> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit X-Virus-Scanned: by amavisd-new at postgresql.org X-Archive-Number: 200310/239 X-Sequence-Number: 3987 Dror, I gave this some serious thought at first. I only deal with int8, numeric(24,12) and varchar(32) columns which I could reduce to 3 different tables. Problem was going from 1700-3000 rows to around 300,000-1,000,000 rows per system per day that is sending data to our database. BTW, the int8 and numeric(24,12) are for future expansion. I hate limits. Greg Dror Matalon wrote: > It's still not quite clear what you're trying to do. Many people's gut > reaction is that you're doing something strange with so many columns in > a table. > > Using your example, a different approach might be to do this instead: > > Day | Name | Value > ------+-------------+----------- > Oct 1 | OS | Solaris 5.8 > Oct 1 | Patch | 108528-12 > Oct 3 | Patch | 108528-13 > > > You end up with lots more rows, fewer columns, but it might be > harder to query the table. On the other hand, queries should run quite > fast, since it's a much more "normal" table. > > But without knowing more, and seeing what the other columns look like, > it's hard to tell. > > Dror -- Greg Spiegelberg Sr. Product Development Engineer Cranel, Incorporated. Phone: 614.318.4314 Fax: 614.431.8388 Email: gspiegelberg@Cranel.com Cranel. Technology. Integrity. Focus.