More attention has been paid to the rapid numerical simulation of wave propagation and transformation in ocean engineering and nearshore hydrodynamics. This paper presents a graphics processing unit (GPU)-accelerated implementation of a high-accuracy two-layer Boussinesq model using Compute Unified Device Architecture. The solution procedure of the original model has been extensively optimized to align with the new parallel computing paradigm, resulting in substantial computational speedups. These optimizations include the restructuring of the solution sequence for the two-layer vertical velocity, which has been reorganized and solved in parallel blocks, as well as the refinement of the tolerance calculation module through atomic operations. Additionally, the cyclic reduction algorithm has been extended using ghost grids and the concatenation matrix method, with three optimization strategies designed for different variables. A series of three-dimensional simulations were conducted to validate the accuracy of the code, the precision of the simulations, and computational efficiency, including deep-water wave group propagation and nearshore wave evolution. The comparison results show that the model can accurately capture the nonlinearity and dispersion of waves from deep to shallow water. For single-GPU computation, compared with serial and parallel central processing unit code, 4.11× and 2.5× speedup is achieved in larger computing grids (2.6×105).
Chen et al. (Tue,) studied this question.