Parallel Semicoarsening Multigrid
Peter N. Brown, Robert D. Falgout and Jim E. Jones

Jim E. Jones

Center for Applied Scientific Computing
Lawrence Livermore National Laboratory
Box 808, L-561
Livermore, CA 94551


Abstract

In this talk we discuss recent results for a 3d semicoarsening multigrid code for solving scalar diffusion problems on structured grids. The algorithm, due to Steve Schaffer, combines operator induced interpolation, semicoarsening and plane relaxation (using a 2d multigrid algorithm to invert the planes). The solver has been tested on diffusion problems and the results demonstrate that the algorithm is robust with respect to jumps and anisotropies in the coefficients. The results also show that the algorithm is scalable, in the sense that its convergence rate does not significantly degrade with problem size. We discuss also the issues related to the parallelization of the our code. The code has been implemented on distributed memory computers and we present results of a scalability study where the problem size per processor is kept constant.