GPU-based Monte Carlo simulation for the design Sea Ice Load

Abstract

Modern Graphics Processing Units (GPUs) with massive number of threads and many-core architectural components support both graphics and general purpose computing. NVIDIA’s compute unified device architecture (CUDA) takes advantage of parallel computing and utilizes the tremendous power of GPUs. The present study demonstrates a high performance computing (HPC) framework for a Monte Carlo simulation to determine design sea ice loads which is then implemented in both GPU and CPU (central processing unit). Results show a speedup of up to 130 times for the 4 Tesla K80 GPUs over an optimized CPU OpenMP (Open Multi-Processing) implementation and a speedup of up to 8 times for the 4 Tesla K80 over a single Tesla K80 GPU implementation. The elapsed time of the different implementations reduced from about 2.5 hours to 0.7 seconds.