Simulating Plasma Turbulence in Tokamaks

Abstract

A challenging and fundamental research problem is the better understanding and control of the turbulent transport of heat in present-day tokamak fusion experiments. Recent developments in numerical methods along with enormous gains in computing power have made large-scale simulations an important tool for improving our understanding of this phenomena. Simulating this highly non-linear behavior requires solving for the perturbations of the phase space distribution function in five dimensions. We use a particle-in-cell approach to solve the equations. The code has been parallelized for a variety of architectures (C90, CM-5, T3D) using a 1-D domain decomposition along the toroidal axis, for which the number of particles in each cell remains approximately constant. The quasi-uniform distribution of particles, which minimizes load imbalance, coupled with the relatively small movement of particles across cells, which minimizes communications, makes this problem ideally suited to massively parallel architectures. We present the performance of the program for different numbers of processors and problem sizes. In addition, we discuss some recent scientific results obtained from the code.