Massively Parallel Fluid Simulations on Amazon's HPC Cloud

In this paper, we report on the results of numerical experiments in the field of computational fluid dynamics (CFD) on Amazon's HPC cloud. To this end, we benchmarked our MPI-parallel fluid solver NaSt3DGPF on several HPC compute nodes of the cloud system. Our solver can use CPUs and GPUs to calculate the simulation results. With a pre-requested number of instances we observe for both, CPUs and GPUs, good scalability even for a larger number of parallel processes, provided that the GPUs run in the non-ECC mode. Furthermore, we see a high potential for medium sized parallel compute problems which are typically present in industrial engineering applications.