Parallel Algorithms / Architecture Synthesis, AIZU International Symposium on (1995)
Aizu-Wakamatsu, Fukushima, Japan
Mar. 15, 1995 to Mar. 17, 1995
T. Rauber , Dept. of Comput. Sci., Saarlandes Univ., Saarbrucken, Germany
G. Runger , Dept. of Comput. Sci., Saarlandes Univ., Saarbrucken, Germany
The spatial discretization of nonlinear partial differential equations (PDEs) results in large systems of nonlinear ordinary differential equations (ODEs). The discretization of the Brusselator equation is a characteristic example. For the parallel numerical solution of the Brusselator equation we use an iterated Runge-Kutta method. We propose modifications of the original method that exploit the access structure of the Brusselator equation. The implementation is realized on an Intel iPSC/860. A theoretical analysis of the resulting speedup values shows that the efficiency cannot be improved considerably.
nonlinear differential equations; partial differential equations; Runge-Kutta methods; iterative methods; parallel algorithms; parallel machines; mathematics computing; distributed solution; Brusselator equation; spatial discretization; nonlinear partial differential equations; nonlinear ordinary differential equations; parallel numerical solution; iterated Runge-Kutta method; access structure; Intel iPSC/860; speedup values
T. Rauber, G. Runger, "Aspects of a distributed solution of the Brusselator equation", Parallel Algorithms / Architecture Synthesis, AIZU International Symposium on, vol. 00, no. , pp. 114, 1995, doi:10.1109/AISPAS.1995.401347