From: | The Hermit Hacker <scrappy(at)hub(dot)org> |
---|---|
To: | Tony Rios <tonester(at)ccom(dot)net> |
Cc: | Shiby Thomas <sthomas(at)cise(dot)ufl(dot)edu>, pgsql-hackers(at)postgreSQL(dot)org |
Subject: | Re: [HACKERS] database size |
Date: | 1998-01-07 00:18:19 |
Message-ID: | Pine.NEB.3.96.980106201741.254c-100000@thelab.hub.org |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
On Tue, 6 Jan 1998, Tony Rios wrote:
> At 06:09 PM 1/6/98 -0500, Shiby Thomas wrote:
>
> >Hi,
>
> >
>
> >I created a table with two columns of type int, and loaded about 300 K records
>
> >in it. So, the total size of the table is approx. that of 600 K integers,
>
> >roughly 2.4 MB.
>
> >But, the file corresponding to the table in pgsql/data/base directory
>
> >has a size of 19 MB. I was wondering if I have done something wrong in
>
> >the installation or usage, or is it the normal behavior ?
>
> >
>
>
> Just wondering.. did you happen to do an INSERT into the database,
>
> then delete some rows.. say 19megs worth, then re-add... From what I've
>
> seen msql db's will always be at least the size of the largest you've ever
>
> had the database before. It will over time, overrite existing deleted
>
> records, but it keeps the data still in there, just sets a delete flag.
>
>
> If you really need to cut the size down, I've had to delete the database
>
> completely, then create another table from scratch. Not sure if there
>
> is a 'purge' type function available, but you have to be careful that
>
> nobody is accessing the db at that time, since it's very sensitive at
>
> that time.
vacuum will clean out the deleted records and truncate the table...has
been so since v6.1, I believe...
Marc G. Fournier
Systems Administrator @ hub.org
primary: scrappy(at)hub(dot)org secondary: scrappy(at){freebsd|postgresql}.org
From | Date | Subject | |
---|---|---|---|
Next Message | The Hermit Hacker | 1998-01-07 00:32:39 | Re: [HACKERS] database size |
Previous Message | Bruce Momjian | 1998-01-06 23:56:52 | Re: consttraints.source |