> Date: Thu, 31 Mar 2005 12:22:00 -0500 > From: Timothy Sloane <[EMAIL PROTECTED]> > Subject: Re: [Prime] Tip: 2^31-1 = Mersenne exponent > To: Michael Vang <[EMAIL PROTECTED]>,
> 2) The Lucas-Lehmer test is sequential, and I've seen it stated that > you can't split up a single exponent across multiple processors in a > multi-CPU machine, so I doubt you could split a single exponent across > multiple computers. Not really. The GLucas program is multi-threaded: it divides the work of the FFT between several threads on several processors on the same computer. IF the amount of data to be shared and exchanged between all threads does not exceed some maximum, GLucas work could be divided between several processes on several machines, using MPI plus a very fast inter-connect like Quadrics. And thus, computing the LTT for a big exponent could be done much faster than on a single mono-processor machine. Tony Acc�dez au courrier �lectronique de La Poste : www.laposte.net ; 3615 LAPOSTENET (0,34�/mn) ; t�l : 08 92 68 13 50 (0,34�/mn) _______________________________________________ Prime mailing list [email protected] http://hogranch.com/mailman/listinfo/prime
