X-Original-To: pgsql-performance-postgresql.org@localhost.postgresql.org Received: from localhost (unknown [200.46.204.2]) by svr1.postgresql.org (Postfix) with ESMTP id 9D3C2D1D861; Wed, 19 Nov 2003 14:11:20 +0000 (GMT) Received: from svr1.postgresql.org ([200.46.204.71]) by localhost (neptune.hub.org [200.46.204.2]) (amavisd-new, port 10024) with ESMTP id 78780-09; Wed, 19 Nov 2003 10:10:53 -0400 (AST) Received: from smtp-send.myrealbox.com (smtp-send.myrealbox.com [192.108.102.143]) by svr1.postgresql.org (Postfix) with ESMTP id 3B30BD1D859; Wed, 19 Nov 2003 10:10:49 -0400 (AST) Received: from myrealbox.com shridhar_daithankar@smtp-send.myrealbox.com [202.54.11.72] by smtp-send.myrealbox.com with NetMail SMTP Agent $Revision: 3.44 $ on Novell NetWare via secured & encrypted transport (TLS); Wed, 19 Nov 2003 07:10:51 -0700 Message-ID: <3FBB79E6.6040701@myrealbox.com> Date: Wed, 19 Nov 2003 19:40:46 +0530 From: Shridhar Daithankar User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.5) Gecko/20031013 Thunderbird/0.3 X-Accept-Language: en-us, en MIME-Version: 1.0 To: josh@agliodbs.com Cc: pgsql-performance@postgresql.org, PostgreSQL-development Subject: Re: More detail on settings for pgavd? References: <200311181558.45847.josh@agliodbs.com> In-Reply-To: <200311181558.45847.josh@agliodbs.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: 200311/157 X-Sequence-Number: 4717 Josh Berkus wrote: > Shridhar, > > I was looking at the -V/-v and -A/-a settings in pgavd, and really don't > understand how the calculation works. According to the readme, if I set -v > to 1000 and -V to 2 (the defaults) for a table with 10,000 rows, pgavd would > only vacuum after 21,000 rows had been updated. This seems wrong. No. that is correct. It is calculated as threshold = base + scale*numebr of current rows Which translates to 21,000 = 1000 + 2*1000 However I do not agree with this logic entirely. It pegs the next vacuum w.r.t current table size which is not always a good thing. I would rather vacuum the table at 2000 updates, which is what you probably want. Furthermore analyze threshold depends upon inserts+updates. I think it should also depends upon deletes for obvious reasons. > Can you clear this up a little? I'd like to tweak these settings but can't > without being better aquainted with the calculation. What did you expected in above example? It is not difficult to tweak pg_autovacuum calculations. For testing we can play around. > Also, you may want to reverse your default ratio for Vacuum/analyze frequency. > True, analyze is a less expensive operation than Vacuum, but it's also needed > less often -- only when the *distribution* of data changes. I've seen > databases where the optimal vacuum/analyze frequency was every 10 min/once > per day. OK vacuum and analyze thresholds are calculated with same formula as shown above but with different parameters as follows. vacthresh = vacbase + vacscale*ntuples anathresh = anabase + anascale*ntuples What you are asking for is vacthresh = vacbase*vacscale anathresh = anabase + anascale*ntuples Would that tilt the favour the way you want? i.e. an analyze is triggered when a fixed *percentage* of table changes but a vacuum is triggered when a fixed *number of rows* are changed. I am all for experimentation. If you have real life data to play with, I can give you some patches to play around. And BTW, this is all brain child of Mathew O.Connor(Correct? I am not good at either names or spellings). The way I wrote pgavd originally, each table got to get separate threshold..:-). That was rather a brute force approach. Shridhar