| From: | Bill Moran <wmoran(at)collaborativefusion(dot)com> |
|---|---|
| To: | Erik Jones <erik(at)myemma(dot)com> |
| Cc: | pgsql-general(at)postgresql(dot)org |
| Subject: | Re: Impact of vacuum full... |
| Date: | 2006-07-21 16:26:02 |
| Message-ID: | 20060721122602.0312df7c.wmoran@collaborativefusion.com |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
On Fri, 21 Jul 2006 10:35:17 -0500
Erik Jones <erik(at)myemma(dot)com> wrote:
> Csaba Nagy wrote:
[snip]
> That is an excellent idea, however, what are the effects of CLUSTER on
> empty tables? Considering that most of our queue tables sit empty until
> their used, our main concern is keep the disk space that they use
> available and our 'cleaning' activities will be done whenever the tables
> empty out (unless they are scheduled for use within, say, an hour) I
> have to wonder at whether or not CLUSTER would do anything to an empty
> table with no actual data to cluster.
Based on that description, have you considered using TRUNCATE to clear out
the tables when you're done using them? Truncate is faster than DELETE
and I believe it's the equivalient of dropping and recreating the table,
which means it will free up the space.
--
Bill Moran
Potential Technologies
http://www.potentialtech.com
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Eric E | 2006-07-21 16:34:24 | Problem getting postmaster PID in pg_regress |
| Previous Message | Scott Marlowe | 2006-07-21 16:20:09 | Re: Impact of vacuum full... |