New 3D Electromagnetic Hybrid Kinetic Adaptive Meshless PICOP Code for Petascale Computer Systems

The new level of complexity associated with the petascale computing era requires the development of new scalable software that can run efficiently on petascale computers and provide analysis, simulation, and prediction of complex phenomena important to the DOE. The recently developed, adaptive and meshless Particle-In-Cloud-Of-Point (PICOP) method is a suitable candidate for the new parallelization requirements, due to its meshless hierarchical structure of overlapping neighborhoods. This project will use a new cluster decomposition approach to apply the new parallelization strategy to petascale computing systems. Commercial Applications and other Benefits as described by the awardee If successful, the proposed approach should provide a breakthrough in 3D kinetic simulations, affecting many important physics problems, such as high-energy-density plasmas, high-Mach-number plasma jet/beam generation, highly intense magnetic fields in dense plasmas, and magneto-inertial fusion.