Re: TPC-C Superdome init.ora ?

Good answer but I do benchmarks so I look to see if the vendors tpcc init.oras have anything, and why they differ in their values/formulae?

1. The first thing is the Superdome can be a 64 way box , so why didn'nt they benchmark that.
2. Is it possible the wrong init.ora was used on the 48 way box?

John

In article <94l926$u6b$1_at_nnrp1.deja.com>,   hailey_kyle_at_my-deja.com wrote:
> I'm unfamiliar with what you are making reference to, but the # of
 CPUs
> is used in some cases to calculate the # of latches protecting some
> structures. Thus the cpu_count can be used to fudge the # of some
> latches. I wouldn't particularly spend too much time studying a TPC-C
> init.ora because often there are cludges which make very little
> difference and no difference at all unless the system is maxed to the
> bleeding point which is not the case of any production system since
> work loads come in waves and any system running as close to the edge
 as
> a TPC that gets a surge will typically go over in any of various ways,
> memory/paging, CPU etc, or just serious reductions in through put due
> to bottle knecks on any of a number of Oracle resources.
>
> Best
> Kyle
>
> In article <94l1mh$nkl$1_at_nnrp1.deja.com>,
> jobrien99_at_my-deja.com wrote:
> > Hi,
> >
> > Just wondering why the cpu_count is set to 64 in the latest tpc c
 with
> > the superdome when there are only 48 cpus ?
> >
> > What about the compatible parameter and the announced version ?
> >
> > John
> >
> > Sent via Deja.com
> > http://www.deja.com/
> >
>
> Sent via Deja.com
> http://www.deja.com/
>

Sent via Deja.com
http://www.deja.com/ Received on Wed Jan 24 2001 - 09:32:53 CST