Re: [PERFORM] function execute on v.9.2 slow down

2013-09-16 Thread Robert Haas
On Mon, Aug 12, 2013 at 8:21 AM, Александр Белинский wrote: > Hi! > I can't explain why function is slow down on same data. > Postgresql.conf the same, hardware is more powerful. > Diffrents is postgresql version Hmm. PostgreSQL 9.2 will sometimes replan queries a number of times where older rel

Re: [PERFORM] stable and immutable functions in GROUP BY clauses.

2013-09-16 Thread Marc Mamin
> > Hello, > > Stable and immutable functions do not improve performance when used within > the GROUP BY clause. > Here, the function will be called for each row. > > To avoid it, I can replace the funtion by its arguments within GROUP BY. Shame on me ! This is of course bullsh... It has not

[PERFORM] How to optimization database for heavy I/O from updates (software and hardware)

2013-09-16 Thread Niels Kristian Schjødt
I'm in the process of taking the next leap in performance optimization of our database, I just need some good advice on my journey. I posted the full question with images here on stackexchange if someone would be interested in commenting / answering it would be great! Regards Niels Kristian ht

Re: [PERFORM] COPY TO and VACUUM

2013-09-16 Thread Roberto Grandi
Hi Jeff, thanks for your suggestion, Well test Vacuum instead of Cluster and come back with live result. at the same time i discovered that our COPY (...) TO are really really slow, I see 0Kb a t the beginning but at the end they grow by 4Kb each second. Our export is standard (i.e.: SELECT a,