Newsgroups: comp.parallel.pvm
From: Graham Edward Fagg <sssfagg@reading.ac.uk>
Subject: Re: random number seeds
Organization: University of Reading, U.K.
Date: Wed, 22 Mar 1995 13:22:46 +0000

Williams et al, mensioned this her paper in Rome's 1994 ePVM conference. The 
solution was for the master (in an SPMD) program to generate the seeds 
for the slaves. As the slaves are on independent machines and the 
generators are themslves repeatable then the start seeds are all you need 
to maintain repeatability. 
Remember if you're doing something like a MonteCarlo simulation across 
multiple machines not to send the same single start seed to the all the 
machines!

Graham.
===============================================================================
Graham Edward Fagg ||| *** Cluster Computing Lab. ***  ||| e-mail me some time
 Computer Science ||  Software Engineering Subject Grp.  ||  G.E.Fagg@rdg.ac.uk
0734-875123 7626 | http://www.cs.reading.ac.uk/people/gef/ | PVM/MPI/LINDA/VMD
===============================================================================



