<span class="var-sub_title">Eulerian Algorithms for the Discretization of Plasma Kinetic Equations</span> SC18 Proceedings

The International Conference for High Performance Computing, Networking, Storage, and Analysis

Eulerian Algorithms for the Discretization of Plasma Kinetic Equations


Student: James L. Juno (University of Maryland)
Supervisor: William Dorland (University of Maryland)

Abstract: While fluid models are common tools in the study of plasmas, many of these systems, whether in astrophysics or the lab, are only weakly collisional and far from equilibrium, making them more accurately described by kinetic equations. Kinetic equations can be computationally demanding due to the need to solve for the distribution function of the particles in a higher dimensional phase space, with position and velocity coordinates. Despite this challenge, the motivation for solving the plasma kinetic equation is large as there remains a vast array of questions concerning collisionless dynamics in real plasma systems. Here we present algorithms in an Eulerian framework for the discretization of the plasma kinetic equation, using a high-order discontinuous Galerkin finite element method due to its arithmetic intensity and parallelizability. Scaling and performance of the algorithm are discussed, and benchmarks of the algorithm are presented as well.

ACM-SRC Semi-Finalist: no

Poster: PDF
Poster Summary: pdf
Reproducibility Description Appendix: PDF


Back to Poster Archive Listing