205	1998	Jarmo Rantakokko	jarmo@tdb.uu.se	Comparison of partitioning strategies for PDE solvers on composite grids

			Abstract

Different partitioning strategies for multiblock grids have been
compared both experimentally and theoretically. The numerical experiments
have been performed on a 512 processor Cray T3D using a compressible two
dimensional Navier-Stokes solver. Some complementary results are obtained
with an advection equation solver using the Cray T3D and also a
Cray T3E-900/LC240. The results show that the behavior of the
different parallelization strategies depends very much on the number
of subgrids and their sizes as well as the number of available
processors. In order to get optimal performance for a certain problem
and processor configuration, the partitioning strategy must be chosen
with regard to these aspects. Our results give guidelines for this.