No abstract

is accepted. If the energy has increased, the move is accepted with a suitable probability. These ideas are shown in the The comllutational load of the Montecarlo-Metropolis flow chart in Fig. 1. algorithm makes the study of parallel computing techniques'very interesting, since advantage can be taken of the powerful low cost processors available nowadays. W e identified some partitioning methods for the problem which give a high degree of processor exploitation, without interfering with the statistical features of the simulation method.In this paper, we present a parallel architecture based on the lntcl i860 RISC-processor and split into two subgroups of processors: the first designed to execute Montecarlo loops, and the second to assess the statistical parameters. This approach enables efficient parallelization because data-transfer recurrence among proccssor groups is relatively small.With regard to the Montecarlo-Metropolis algorithm we expect to reach performances higher than those given b y present Scneration of CRAYs.