COLLOQUIUM Department of Computer Science and Engineering University of South Carolina Parallel Particle Tracking Methods for Scientific Computing Jing-Ru (Ruth) C. Cheng* Department of Computer Science and Engineering Pennsylvania State University Date: February 18, 2002 (Monday) Time: 3:30-4:30PM Place: 300 Main B103 Abstract Particle tracking methods are a versatile computational technique central to the simulation of a wide range of scientific applications including: visualization, molecular dynamics, direct simulation Monte Carlo methods, and Eulerian-Lagrangian methods. We introduce a common framework, the "in-element" particle tracking method, based on the assumption that particle trajectories are computed by problem data localized to individual elements. The goal of the software implementation is to design the interface to be portable between different programming environments, while the component interface is lightweight and functional---allowing for easy incorporation by other parallel programming environments. This colloquium presents a new parallel approach for the dynamic partitioning of particle-mesh computational systems, and demostrates a new a posteriori error estimator based on particle tracking methods for adaptive mesh refinement. Experiment results detail the performance of this parallel load balancing particle method for two- and three-dimensional particle-mesh test problems on an unstructured, adaptive mesh. Ruth Cheng received her first Ph.D. in Civil and Environmental Engineering from the Pennsylvania State University in 1995. Before 1998, the main focus of her research was on model development of subsurface flow and sediment transport. As a Ph.D. candidate in Computer Science and Engineering at Penn State since 1998, her research interests have been in parallel computation and software development including large-scale scientific computing, parallel algorithms and object-oriented programming methods. Additionally, she still occasionally works on modeling reactive transport. She has been in the Mathematics and Computer Science Division at Argonne National Laboratory as a Given's Fellow for two summers. Her research at Argonne with the PETSc team involved a new discretization technique. Currently, she is an NSF Graduate Fellow through NSF's IGERT program. Her Ph.D. dissertation is entitled "Efficient Parallelization of In-Element Particle Tracking Methods." In the near future, she would like to extend her research to component-based scientific computing and scientific visualization for developing virtual environments used to visualize three-dimensional numerical results. *Dr. Cheng is a faculty candidate.