Re: performance problem - 10.000 databases - Mailing list pgsql-admin

From scott.marlowe
Subject Re: performance problem - 10.000 databases
Date
Msg-id Pine.LNX.4.33.0311051403510.10816-100000@css120.ihs.com
Whole thread Raw
In response to Re: performance problem - 10.000 databases  (Marek Florianczyk <franki@tpi.pl>)
Responses Re: performance problem - 10.000 databases  (Marek Florianczyk <franki@tpi.pl>)
List pgsql-admin
On 5 Nov 2003, Marek Florianczyk wrote:

> W li¶cie z ¶ro, 05-11-2003, godz. 19:52, Tom Lane pisze:
> > Marek Florianczyk <franki@tpi.pl> writes:
> > > Maybe reconnect is to often, but how to explain that reular queries like
> > > select * from table1 ale much faster than \d's ? ( my post to Jeff )
> >
> > [ further experimentation... ]  Ah-hah, I see the problem in 7.3, though
> > not in 7.4 which is what I was testing to begin with.  7.4 is smarter
> > about optimizing the LEFT JOINs that are used in \d's queries.
> >
>
> So how do you think
> sould I try v7.4 on production machine and wait for stable version?
> Or better use v7.3 with 3000 databases?
> I have to do that till december.
> Will this optimization solve problem "\d" queries, or just speed it up
> few seconds ?

7.4 just went RC1 and is looking good for release in <2 weeks.  I'd test
it to see if it works for you.  I've found the .0 releases in postgresql
to be good enough for production, even if they do have one or two minor
bugs that pop up.  Take frequent backups (you should anyway) and you'll
probably be alright.

Heck, you're already pushing the performance envelope with 3,000 users,
might as well go for the faster of the two and you'll have one less
scheduled upgrade ahead of you.

When do you need to go live?  If it's >1 month, then I'd definitely
recommend 7.4.


pgsql-admin by date:

Previous
From: Tom Lane
Date:
Subject: Re: why can't select for update be used with join
Next
From: Juan Miguel
Date:
Subject: Re: Microsoft access verses postgresql