On 14 Oct 99, at 18:14, St. Dee wrote:
> I'm running mprime (v19) on a dual-processor box (RH 6.0, very basic, no
> graphical interface at installed) and am curious about the hit others take
> when moving from using one to two processors. (People running duals under
> NT are also welcomed to respond!)
I'm running two dual systems, both running NT WS 4.0 SP5. One has 2 x
PII-350 and one has 2 x PIII-450.
>
> If I run a single instance of mprime, I get LL iteration times on exponents
> near 8,200,000 of about .220. If I run two instances of mprime, each gets
> iteration times of around .245. I expected some hit, but I have no idea if
> that is too big of a hit or not.
Sounds about right. The performance hit gets bigger with bigger clock
multipliers. The problem is that having 2 processors accessing 1
memory bus tends to cause memory bus congestion.
> Curiously, I did notice that when the box
> was doing some factoring to 64 bits, it didn't seem to make any difference
> in the factoring times whether I had one or two processors crunching.
Yes. Trial factoring will run from the processor cache (even on a
Celeron) so doesn't hit the memory bus. Ideally you want to be
running something different on the processors of a dual system; a mix
of ECM on small exponents & LL testing also works well.
>
> In case it matters, the box contains 2 Celeron 466 processors on an Abit
> BP6 board. Anyone looking for a big speed bump at low cost, try out a
> combo like this!
I have it on good authority that Intel are trying hard to prevent
people constructing dual Celeron systems - because of the potential
damage to the PIII market. Be warned, you might buy the bits & it
might not work 8-(
Regards
Brian Beesley
_________________________________________________________________
Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm
Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
