Re: [HACKERS] Slow count(*)

On 02/01/2008, Abraham, Danny <danny_abraham(at)bmc(dot)com> wrote:
> When comparing the OLTP part of our product, PG is about 15% slower
> compared to Oracle, which is reasonable.
>
> When comparing the DSS part, it is about 20 times slower.
>
> Unfortunately, we need both.
>

Send slow queries, please, and than we can help you.

don't forget on VACUUM and ANALYZE statements before and then send
execution plans

http://www.postgresql.org/docs/8.2/static/sql-explain.html

>
> -----Original Message-----
> From: Pavel Stehule [mailto:pavel(dot)stehule(at)gmail(dot)com]
> Sent: Wednesday, January 02, 2008 5:47 PM
> To: Abraham, Danny
> Cc: pgsql-hackers(at)postgresql(dot)org
> Subject: Re: [HACKERS] Slow count(*)
>
> On 02/01/2008, Abraham, Danny <danny_abraham(at)bmc(dot)com> wrote:
> > Hi,
> >
> > We are looking for a patch that will help us count using the indexes.
> >
> > Our product is about 20 times slower on Postgres compared to MS SQL
> > Server.
> >
> > Any ideas?
>
> There isn't any similar patch and will not be.
>
> Use materialized views or similar techniques.
>
> Are you sure, so all your problems are only in SELECT COUNT(*)?
>
> Check, please, all slow queries.
>
> Regards
> Pavel Stehule
> >
> > Danny Abraham
> > BMC Software
> > CTM&D Business Unit
> > 972-52-4286-513
> > danny_abraham(at)bmc(dot)com
> >
> >
> > ---------------------------(end of
> broadcast)---------------------------
> > TIP 5: don't forget to increase your free space map settings
> >
>
>