Parallel computation of a dam-break flow model using OpenMP on a multi-core computer

Abstract

High-performance calculations are of great importance to the simulation of dam-break events, as discontinuous solutions and accelerated speed are key factors in the process of dam-break flow modeling. In this study, Roe’s approximate Riemann solution of the finite volume method is adopted to solve the interface flux of grid cells and accurately simulate the discontinuous flow, and shared memory technology (OpenMP) is used to realize parallel computing. Because an explicit discrete technique is used to solve the governing equations, and there is no correlation between grid calculations in a single time step, the parallel dam-break model can be easily realized by adding OpenMP instructions to the loop structure of the grid calculations. The performance of the model is analyzed using six computing cores and four different grid division schemes for the Pangtoupao flood storage area in China. The results show that the parallel computing improves precision and increases the simulation speed of the dam-break flow, the simulation of 320 h flood process can be completed within 1.6 h on a 16-kernel computer; a speedup factor of 8.64× is achieved. Further analysis reveals that the models involving a larger number of calculations exhibit greater efficiency and a higher rate of acceleration. At the same time, the model has good extendibility, as the speedup increases with the number of processor cores. The parallel model based on OpenMP can make full use of multi-core processors, making it possible to simulate dam-break flows in large-scale watersheds on a single computer.