DATA-PARALLEL LOWER-UPPER RELAXATION METHOD FOR REACTING FLOWS

The implicit lower-upper symmetric Gauss-Seidel (LU-SGS) method of Yoo n and Jameson is modified for use on massively parallel computers. The method has been implemented on the Thinking Machines CM-5 and the Mas Par MP-1 and MP-2, where large percentages of the theoretical peak flo ating point performance are obtained. It is shown that the new data-pa rallel LU relaxation method has better convergence properties than the original method for two different inviscid compressible flow simulati ons. The convergence is also improved for five-species reacting air co mputations. The performance of the method on various partitions of the CM-5 and on the MasPar computers is discussed. The new method shows p romise for the efficient simulation of very large perfect gas and reac ting flows.