Re: For index bloat: VACUUM ANALYZE vs REINDEX/CLUSTER - Mailing list pgsql-general

From Vivek Khera
Subject Re: For index bloat: VACUUM ANALYZE vs REINDEX/CLUSTER
Date
Msg-id 6D316CC5-9ED8-4479-A6BE-75DE6D0F9DB4@khera.org
Whole thread Raw
In response to Re: For index bloat: VACUUM ANALYZE vs REINDEX/CLUSTER  ("Phoenix Kiula" <phoenix.kiula@gmail.com>)
Responses Re: For index bloat: VACUUM ANALYZE vs REINDEX/CLUSTER  (Gregory Stark <stark@enterprisedb.com>)
Re: For index bloat: VACUUM ANALYZE vs REINDEX/CLUSTER  ("Phoenix Kiula" <phoenix.kiula@gmail.com>)
List pgsql-general
On Sep 24, 2007, at 12:00 PM, Phoenix Kiula wrote:

> I feel your pain. But I seem to have (mostly) solved my problem in
> three ways:

My particular usage pattern (add data continuously, purge *some* of
the data once per week or every other week.  The purge is what seems
to kill it.  Last time I reindexed, I reclaimed over 20Gb of disk
space.  That was after 6 months from the prior reindex.

Recommending I run vacuum intermixed with the data purge is a non-
starter; the vacuum on these tables takes a couple of hours.  I'd
never finish purging my data with that kind of delay.

Recommending splitting my tables with inheritance is not going to
work since my purges don't happen across any lines which would make
sense for splitting with inheritance.

I will investigate the fill-factor.  That seems like it may make some
sense the way I do inserts and updates...


pgsql-general by date:

Previous
From: Erik Jones
Date:
Subject: Re: pg_dumping large objects
Next
From: Alvaro Herrera
Date:
Subject: Re: pg_dumping large objects