Parallel molecular dynamics for silicon and silicon carbide: MPI, CUDA and CUDA-MPI implementation

In this paper, the creation of a hybrid computer code based on the joint use of CUDA and MPI technology for silicon Si and silicon carbide SiC is considered in the framework of the molecular dynamics method. The efficiency of the created code is compared both with a program that uses only the MPI interface, and with a CUDA program designed to work only with one GPU.