Developing Highly Scalable Particle-Mesh Codes for GPUs: A Generic Approach


Developing Highly Scalable Particle-Mesh Codes for GPUs: A Generic Approach

Juckeland, G.; Bussmann, M.

Dive deep into a multi-parallel Particle in Cell code that utilizes MPI, pthreads, and CUDA. Around this specific application a general C++ framework for transparent data transfers between GPUs has been developed and will be presented. Further techniques employed include interleaving of communication and computation, particle tiling and a study of how well CUDA performance can be transferred to OpenCL.

Keywords: gpu; particle-in-cell; pic; algorithm; cuda; simulation; performance; current deposition; cluster; mpi; opencl

Permalink: https://www.hzdr.de/publications/Publ-14546
Publ.-Id: 14546