Parallel Algorithms of Well-Balanced and Weighted Average Flux for Shallow Water Model Using CUDA

Figure 10

Registers per thread and occupancy on a global kernel with __launch_bounds__.