Received: from localhost (wm.hub.org [200.46.204.128]) by postgresql.org (Postfix) with ESMTP id 8B9029FB357 for ; Thu, 28 Sep 2006 14:53:35 -0300 (ADT) Received: from postgresql.org ([200.46.204.71]) by localhost (mx1.hub.org [200.46.204.128]) (amavisd-new, port 10024) with ESMTP id 68214-06 for ; Thu, 28 Sep 2006 17:53:31 +0000 (UTC) X-Greylist: from auto-whitelisted by SQLgrey- Received: from floppy.pyrenet.fr (news.pyrenet.fr [194.116.145.2]) by postgresql.org (Postfix) with ESMTP id DD0539FB361 for ; Thu, 28 Sep 2006 14:53:30 -0300 (ADT) Received: by floppy.pyrenet.fr (Postfix, from userid 106) id 3D4EC30E06; Thu, 28 Sep 2006 19:53:29 +0200 (MET DST) From: "Carlo Stonebanks" X-Newsgroups: pgsql.performance Subject: Re: Performace Optimization for Dummies Date: Thu, 28 Sep 2006 13:53:22 -0400 Organization: Hub.Org Networking Services Lines: 31 Message-ID: References: X-Complaints-To: usenet@news.hub.org X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 6.00.2900.2869 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.2962 X-RFC2646: Format=Flowed; Response To: pgsql-performance@postgresql.org X-Virus-Scanned: Maia Mailguard 1.0.1 X-Spam-Status: No, hits=0 tagged_above=0 required=5 tests= X-Spam-Level: X-Archive-Number: 200609/391 X-Sequence-Number: 21077 > are you using the 'copy' interface? Straightforward inserts - the import data has to transformed, normalised and de-duped by the import program. I imagine the copy interface is for more straightforward data importing. These are - buy necessity - single row inserts. > thats a tough question. my gut says that windows will not scale as > well as recent linux kernels in high load environments. But not in the case of a single import program trying to seed a massive database? > hearing good things about the woodcrest. pre-woodcrest xeon (dempsey > down) is outclassed by the opteron. Need to find a way to deterimine the Xeon type. The server was bought in early 2006, and it looks like woodcrest was form July. > 1. can probably run fsync=off during the import > 2. if import is single proecess, consider temporary bump to memory for > index creation. or, since you have four cores consider having four > processes import the data somehow. > 3. turn off stats collector, stats_command_string, stats_row_level, > and autovacuum during import. Very helpful, thanks. Carlo