Parallelization of plasma simulation codes: gridless finite size particle versus particle in cell approach

Abstract

The main features of gridless finite-size-particle codes are examined and compared with those of particle-in-cell ones, from the point of view of the performances that can be obtained, with respect to both the spatial-resolution level and the efficiency of parallel particle simulations. The particle-decomposition parallelization strategy is discussed, and its implementation in the framework of the High Performance Fortran paradigm is presented. It is shown that such codes are particularly suited for particle-decomposition parallelization on distributed-memory architectures, as they present a strong reduction, in comparison with particle-in-cell codes, of the memory and computational offsets related to storing and updating the replicated fluctuating-field arrays.