Re: [HACKERS] More detail on settings for pgavd?
Robert Treat <xzilla@users.sourceforge.net>
From: Robert Treat <xzilla@users.sourceforge.net>
To: "Matthew T. O'Connor" <matthew@zeut.net>
Cc: Josh Berkus <josh@agliodbs.com>, Shridhar Daithankar <shridhar_daithankar@myrealbox.com>, pgsql-performance@postgresql.org, PostgreSQL-development <pgsql-hackers@postgresql.org>
Date: 2003-11-21T14:14:14Z
Lists: pgsql-hackers, pgsql-performance
On Thu, 2003-11-20 at 19:40, Matthew T. O'Connor wrote:
> I'm open to discussion on changing the defaults. Perhaps what it would
> be better to use some non-linear (perhaps logorithmic) scaling factor.
> So that you wound up with something roughly like this:
>
> #tuples activity% for vacuum
> 1k 100%
> 10k 70%
> 100k 45%
> 1M 20%
> 10M 10%
> 100M 8%
>
Just thinking out loud here, so disregard if you think its chaff but...
if we had a system table pg_avd_defaults that held what we generally
consider the best default percentages based on reltuples/pages, and
added a column to pg_class (could be some place better but..) which
could hold an overriding percentage, you could then have a column added
to pg_stat_all_tables called vacuum_percentage, which would be a
coalesce of the override percentage or the default percentages based on
rel_tuples (or rel_pages). This would give autovacuum a place to look
for each table as to when it should vacuum, and gives administrators the
option to tweak it on a per table basis if they find they need a
specific table to vacuum at a different rate than the "standard".
Robert Treat
--
Build A Brighter Lamp :: Linux Apache {middleware} PostgreSQL