On 6 Feb 00, at 12:50, Jason Stratos Papadopoulos wrote:
> > > However, the FFT itself is very amenable to parallel processing
> > > techniques - on a processor with N independent compute pathways, you
> > > can compute N elements in the same time that a single element would
> > > take to compute just one.
> > >
> > How much of the mersenne-test is used for the FFT then?
> > Isn't the FFT the single most time-consuming part?
> >
> > Wouldn't it make sense to make a parallelized version of prime95 that
> > distributes the FFT over N processors?
The trouble here is that you need vector processors - not SMP - to do
the job properly i.e. all the processors share memory so there is not
really a communication problem.
Actually we already do a bit of interleaving - even on uniprocessor
systems - to make best use of the processor pipeline & cache line
characteristics.
Regards
Brian Beesley
_________________________________________________________________
Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm
Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
