RE: v7.1b4 bad performance

Okay, am going to guess that all four of these results are with
CommitDelay=0, and for each higher number of clients, the results get
better and better ... what if you increase CommitDelay back to 5, for
those various levels of client connections?

On Fri, 16 Feb 2001, Schmidt, Peter wrote:

>
>
> > BTW, have you tried pgbench with multiple clients (-c) rather
> > than just
> > one?
> >
> > regards, tom lane
>
> I was wrong about multiple clients...
> I've seen as many as 230 tps with CommitDelay=0.
> Peter
>
> number of clients: 100
> number of transactions per client: 10
> number of transactions actually processed: 1000/1000
> tps = 47.763389(including connections establishing)
> tps = 50.660709(excluding connections establishing)
>
> number of clients: 50
> number of transactions per client: 20
> number of transactions actually processed: 1000/1000
> tps = 76.521240(including connections establishing)
> tps = 79.924979(excluding connections establishing)
>
> number of clients: 10
> number of transactions per client: 100
> number of transactions actually processed: 1000/1000
> tps = 103.444345(including connections establishing)
> tps = 104.723556(excluding connections establishing)
>
> number of clients: 1
> number of transactions per client: 1000
> number of transactions actually processed: 1000/1000
> tps = 172.223961(including connections establishing)
> tps = 172.774743(excluding connections establishing)
>

Marc G. Fournier ICQ#7615664 IRC Nick: Scrappy
Systems Administrator @ hub.org
primary: scrappy(at)hub(dot)org secondary: scrappy(at){freebsd|postgresql}.org