pg.ddx.io pgsql-performance@postgresql.org mailing list archive
help / color / mirror / Atom feedRe: Réf
4+ messages / 3 participants
[nested] [flat]
* Re: Réf
@ 2005-04-06 17:42 Steinar H. Gunderson <sgunderson@bigfoot.com>
0 siblings, 1 reply; 4+ messages in thread
From: Steinar H. Gunderson @ 2005-04-06 17:42 UTC (permalink / raw)
To: pgsql-performance
On Wed, Apr 06, 2005 at 01:18:29PM -0400, Rod Taylor wrote:
> Yeah, I think that can be done provided there is more than one worker.
> My limit seems to be about 1000 transactions per second each with a
> single insert for a single process (round trip time down the Fibre
> Channel is large) but running 4 simultaneously only drops throughput to
> about 900 per process (total of 2400 transactions per second) and the
> machine still seemed to have lots of oomph to spare.
Erm, have I missed something here? 900 * 4 = 2400?
/* Steinar */
--
Homepage: http://www.sesse.net/
^ permalink raw reply [nested|flat] 4+ messages in thread
* Re: Réf
@ 2005-04-06 18:23 Rod Taylor <pg@rbt.ca>
parent: Steinar H. Gunderson <sgunderson@bigfoot.com>
0 siblings, 1 reply; 4+ messages in thread
From: Rod Taylor @ 2005-04-06 18:23 UTC (permalink / raw)
To: Steinar H. Gunderson <sgunderson@bigfoot.com>; +Cc: pgsql-performance
On Wed, 2005-04-06 at 19:42 +0200, Steinar H. Gunderson wrote:
> On Wed, Apr 06, 2005 at 01:18:29PM -0400, Rod Taylor wrote:
> > Yeah, I think that can be done provided there is more than one worker.
> > My limit seems to be about 1000 transactions per second each with a
> > single insert for a single process (round trip time down the Fibre
> > Channel is large) but running 4 simultaneously only drops throughput to
> > about 900 per process (total of 2400 transactions per second) and the
> > machine still seemed to have lots of oomph to spare.
>
> Erm, have I missed something here? 900 * 4 = 2400?
Nope. You've not missed anything.
If I ran 10 processes and the requirement would be met.
--
^ permalink raw reply [nested|flat] 4+ messages in thread
* Re: Réf
@ 2005-04-06 18:40 Alex Turner <armtuk@gmail.com>
parent: Rod Taylor <pg@rbt.ca>
0 siblings, 1 reply; 4+ messages in thread
From: Alex Turner @ 2005-04-06 18:40 UTC (permalink / raw)
To: Rod Taylor <pg@rbt.ca>; +Cc: Steinar H. Gunderson <sgunderson@bigfoot.com>; pgsql-performance
I think his point was that 9 * 4 != 2400
Alex Turner
netEconomist
On Apr 6, 2005 2:23 PM, Rod Taylor <pg@rbt.ca> wrote:
> On Wed, 2005-04-06 at 19:42 +0200, Steinar H. Gunderson wrote:
> > On Wed, Apr 06, 2005 at 01:18:29PM -0400, Rod Taylor wrote:
> > > Yeah, I think that can be done provided there is more than one worker.
> > > My limit seems to be about 1000 transactions per second each with a
> > > single insert for a single process (round trip time down the Fibre
> > > Channel is large) but running 4 simultaneously only drops throughput to
> > > about 900 per process (total of 2400 transactions per second) and the
> > > machine still seemed to have lots of oomph to spare.
> >
> > Erm, have I missed something here? 900 * 4 = 2400?
>
> Nope. You've not missed anything.
>
> If I ran 10 processes and the requirement would be met.
> --
>
> ---------------------------(end of broadcast)---------------------------
> TIP 4: Don't 'kill -9' the postmaster
>
^ permalink raw reply [nested|flat] 4+ messages in thread
* Re: Réf
@ 2005-04-06 19:01 Rod Taylor <pg@rbt.ca>
parent: Alex Turner <armtuk@gmail.com>
0 siblings, 0 replies; 4+ messages in thread
From: Rod Taylor @ 2005-04-06 19:01 UTC (permalink / raw)
To: Alex Turner <armtuk@gmail.com>; +Cc: Steinar H. Gunderson <sgunderson@bigfoot.com>; pgsql-performance
On Wed, 2005-04-06 at 14:40 -0400, Alex Turner wrote:
> I think his point was that 9 * 4 != 2400
Oh.. heh.. I didn't even notice that.
Can I pretend I did it in my head using HEX math and that it wasn't a
mistake?
> On Apr 6, 2005 2:23 PM, Rod Taylor <pg@rbt.ca> wrote:
> > On Wed, 2005-04-06 at 19:42 +0200, Steinar H. Gunderson wrote:
> > > On Wed, Apr 06, 2005 at 01:18:29PM -0400, Rod Taylor wrote:
> > > > Yeah, I think that can be done provided there is more than one worker.
> > > > My limit seems to be about 1000 transactions per second each with a
> > > > single insert for a single process (round trip time down the Fibre
> > > > Channel is large) but running 4 simultaneously only drops throughput to
> > > > about 900 per process (total of 2400 transactions per second) and the
> > > > machine still seemed to have lots of oomph to spare.
> > >
> > > Erm, have I missed something here? 900 * 4 = 2400?
> >
> > Nope. You've not missed anything.
> >
> > If I ran 10 processes and the requirement would be met.
> > --
> >
> > ---------------------------(end of broadcast)---------------------------
> > TIP 4: Don't 'kill -9' the postmaster
> >
>
--
^ permalink raw reply [nested|flat] 4+ messages in thread
end of thread, other threads:[~2005-04-06 19:01 UTC | newest]
Thread overview: 4+ messages (download: mbox mbox.gz follow: Atom feed)
-- links below jump to the message on this page --
2005-04-06 17:42 Re: Réf Steinar H. Gunderson <sgunderson@bigfoot.com>
2005-04-06 18:23 ` Rod Taylor <pg@rbt.ca>
2005-04-06 18:40 ` Alex Turner <armtuk@gmail.com>
2005-04-06 19:01 ` Rod Taylor <pg@rbt.ca>
This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox