| From: | Tino Wildenhain <tino(at)wildenhain(dot)de> |
|---|---|
| To: | Guido Neitzer <guido(dot)neitzer(at)pharmaline(dot)de> |
| Cc: | pgsql-general General <pgsql-general(at)postgresql(dot)org> |
| Subject: | Re: Slow COUNT |
| Date: | 2005-12-04 13:40:49 |
| Message-ID: | 1133703649.5734.75.camel@Andrea.peacock.de |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
Am Sonntag, den 04.12.2005, 14:02 +0100 schrieb Guido Neitzer:
> On 02.12.2005, at 20:02 Uhr, Jaime Casanova wrote:
>
> > so the way to do it is create a trigger that record in a table the
> > number of rows...
>
> As there are SO MANY questions about the "count(*)" issue, I wonder
> whether it makes sense to add a mechanism which does exactly the
> method mentioned above in a default PostgreSQL installation (perhaps
> switched of by default for other performance impacts)?!
I dont think this would match postgres style - to include
a kludge for a rarely usefull special case. I may be wrong
but personally I never needed unqualified count(*) on a
table to be very fast.
Doing something to enable aggregates in general to use
an existent index would be a nice ide imho.
(With all the visibility hinting in place)
Just my 0.02Ct.
++Tino
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Martijn van Oosterhout | 2005-12-04 14:23:19 | Re: int to inet conversion |
| Previous Message | Guido Neitzer | 2005-12-04 13:02:10 | Re: Slow COUNT |