Adaptive Runtime Support for Direct Simulation Monte Carlo Methods on Distributed Memory Architectures

Abstract

In highly adaptive irregular problems such as many Particle-In-Cell (PICJ codes and Dimet Simulation Monte Carlo (DSMCJ codes, data access patterns may vary from time step to time step. This fluctuation may hinder efficient utilization of distributed memory parallel computers because of the resulting overhead for data redistribution and dynamic load balancing. To efficiently parallelize such adaptive irregular problems on distributed memory parallel computers, several issues such as effective methods for domain partitioning and fast data transportation must be addressed. This paper presents efficient runtime support methods for such problems. A simple one-dimensional domain partitioning method is implemented and compared with unstructured mesh partitioners such as recursive coordinate bisection and recursive inertial bisection. A remapping decision policy has been investigated for dynamic load balancing on S-dimensional DSMC codes. Performance results are presented (Also cross-referenced as UMIACS-TR-95-27)