20080930, 09:23  #1 
3364_{10} Posts 
Parallel version of Prime for HPC?
Does anyone know of a parallel implementation of prime search for distributed memory clusters. I tinkered around with building one a while back, but time isn't on my side. I know it would be a large undertaking to create an MPI version and post it to GIMPS, but I know of many HPC centers where large scale machines site idle part of the time. I could dedicate a couple million CPU hours to the project if there was a parallel version. I have found some MPI versions of FFT, but I'm not a mathematician, I'm a HPC system admin with a background in hardware, a BS in CS, but as my teachers would tell you, I'm not that great at mathematics! I like to reply back "a computer can add, and all the rest if just decompositions of that".
Thanks for the help,  Donald Tripp University of Hawaii, Hilo 
20080930, 09:56  #2 
Aug 2003
Europe
302_{8} Posts 

20080930, 11:12  #3  
Jan 2008
France
2^{2}·149 Posts 
Quote:
IIRC Ernst wrote somewhere on this forum that he would release a newer parallel version of his Mlucas program, the one that was used to doublecheck the 2 latest primes found. But my memory has been proven wrong enough that you should not trust me blindly :) Last fiddled with by ldesnogu on 20080930 at 11:15 

20080930, 19:52  #4  
∂^{2}ω=0
Sep 2002
República de California
5×2,351 Posts 
Quote:


Thread Tools  
Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
Prime Gap Search latest version of the c code  pinhodecarlos  Prime Gap Searches  170  20191210 19:33 
What's the best way to use PFGW in parallel?  CRGreathouse  Information & Answers  4  20160306 00:21 
Which bits of gmpecm are now parallel?  fivemack  GMPECM  5  20100905 06:49 
Parallel memory bandwidth  fivemack  Factoring  14  20080611 20:43 
Parallel Prime Search  DonaldTripp  Software  2  20070217 19:35 