| From: | Ian Barwick <barwick(at)gmail(dot)com> |
|---|---|
| To: | David Parker <dparker(at)tazznetworks(dot)com> |
| Cc: | pgsql-general(at)postgresql(dot)org |
| Subject: | Re: estimating table size |
| Date: | 2004-07-26 22:39:06 |
| Message-ID: | 1d581afe04072615397222824f@mail.gmail.com |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
On Mon, 26 Jul 2004 18:14:06 -0400, David Parker
<dparker(at)tazznetworks(dot)com> wrote:
> Given a table, foo, created in a database but not populated, is there a
> procedure that will return an estimate of the size of a given tuple in
> that table? It looks like pgstattuple reports on actual table pages; I'm
> looking for something that reads the lengths of each row, and knows what
> the storage overhead would be (extra 4 bytes for varchar, for instance).
>
> Alternatively, is there documentation somewhere that can give me a good
> rule of thumb for this? I can add up the lengths of the columns and make
> my own judgements about averages for varchar columns. The "hidden"
> storage overhead is what I'm wondering about, I guess.
There is a little info in the FAQ:
http://www.postgresql.org/docs/faqs/FAQ.html
particularly sections 4.6 and 4.14
Ian Barwick
barwick(at)gmail(dot)com
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Tom Lane | 2004-07-27 01:00:26 | Re: vacuumdb hanging database cluster |
| Previous Message | Igor Shevchenko | 2004-07-26 22:29:11 | parsing binary varchar[]-s |