Design and implementation issues of parallelizing atmospheric models

摘要

In this paper we present a parallel implementation of the numerical weather prediction Eta model on distributed memory MIMD processors. The required interprocessor communication was implemented and validated on the message passing library Message Passing Interface (MPI). Domain decomposition techniques are used for the horizontal discretization, whereas in the vertical direction each column per grid point is computed. It is found that for a mesh network topology the communication to computation ratio is minimized if the computation domain of each processor is a square of side {square root of}(N/p) for a {square root of}N×{square root of}N mesh of grid points, where p is the number of processors. Moreover, we show that the number of grid points per processor plays an important role in the communication complexity and when t_s < t_ω{square root of}(N/p), where t_s is the start-up time and t_ω is the per word transfer time, then the communication to computation ratio is improved by an order of magnitude. All data communication for finite differencing and I/O is encapsulated within a parallel library. Finally, we present performance results on the Convex Exemplar SPP-2000.