PARALLELIZATION OF DIRECT SIMULATION MONTE-CARLO METHOD COMBINED WITHMONOTONIC LAGRANGIAN GRID

The monotonic Lagrangian grid (MLG) and the direct simulation Monte Ca rlo (DSMC) methodology were combined on the Thinking Machines CM-5 to create a fast DSMC-MLG code with automatic grid adaptation based on lo cal number densities. The MLG is a data structure in which particles t hat are close in physical space are also close in computer memory, Usi ng the MLG data structure, physical space is divided into a number of templates (cells), each containing the same number of particles, An ML G-regularization method, stochastic grid restructuring, is implemented to minimize the occurrence of highly skewed cells, Parallelization of the DSMC-MLG is achieved by two different mapping techniques, First, simulated particles are mapped onto the parallel processors for the pa rticle-oriented processes, such as convection, boundary interactions, and MLG sorting, Second, particle templates are mapped onto the proces sors for computing the macroscopic quantities (i.e., pressure, velocit y, density, and temperature) and statistical sampling, In both levels of mapping, the code logic focuses on the structured and fast communic ations on the CM-5 architecture, The computing time required by the pa rallel DSMC-MLG code was significantly decreased compared with other p arallel efforts and its parallel efficiency on 512 processors achieved approximately 80% for simulation involving one-half million particles .