Nonlinear dynamic analysis efficiency by using a GPU parallelization

A graphics processing unit (GPU) parallelization approach was implemented to improve the efficiency of nonlinear dynamic analysis. The GPU parallelization approach speeded up the computation of implicit time integration and reduced total calculation time. In addition, a parallel equations solver is introduced to solve the equation system. Numerical examples of reinforced concrete (RC) frames were used to investigate the parallel computing speedup of the GPU parallelization approach. An implementation of these RC frame models for fiber beam-column elements was presented. The parallel finite element program is developed to provide parallel execution on personal computer (PC) with different CUDA-capable GPUs. The different number of degrees of freedom from low to high was adopted in the numerical examples. Detailed tests on accuracy, runtime, and speedup are conducted on different GPUs. The nonlinear dynamic response using the GPU parallelization program was in good agreement with that obtained by ABAQUS. Numerical studies indicate that compared with original sequential approach, the GPU parallelization program achieves a 22 times speedups of the solving equation system and improves the overall efficiency of time integration by up to 94%.