Re: about virtual processors

Art - I can't tell you how giggly I am that you seem to be coming around to my long held assertion that the "conventional wisom" of: NUMCPUVPS = realCPUs-1 (or so) ain't necessarily so.....could it be your switch to Sun/Solaris that's changing your mind?? Most of my (admittedly limited) testing along this line has been Solaris based. Have you really *empirically* observed 2 CPUVP's on a 2 CPU machine producing less throughput than 1 CPUVP on a 2 CPU machine?? Even on Solaris?? I still can't imagine that the overhead could be that great.... Greg Bloomberg L.P wrote: > > For a 2CPU system most of us recommend using only a single CPU VP. Based on > this, set: > > MULTIPROCESSOR 1 > NUMCPUVPS 1 > SINGLE_CPU_VP 1 > NOAGE 1 > AFF_SPROC 0 > AFF_NPROCS 0> > This is because IDS does not take advantage of only two CPUs well due to the > overhead of managing the additional VPs. With 4 or more CPUs most of us use, > > and recommend using, one less CPU VP than CPU. However, there is some hint > that faster CPUs, ie >300MHZ MAY be able to provide enough horsepower to add > more than one CPU VP per processor which would imply that 2 CPU VPs on your > system MAY gain you something. I am experimenting with 30 CPU VPs on a 20 > CPU Sun ES6000 w/ 360MHZ processors. YMMV so experiment. snip.....