ABSTRACT
In this paper a parallelization of a Shallow Water numerical scheme suitable for Graphics Processor Unit (GPU) architectures under the NVIDIATM ‘s Compute Unified Device Architecture (CUDA) framework is presented. In order to provide robust, fast and accurate simulations of real flood events, the system features a state-of-the-art Finite Volume explicit discretization technique which is well balanced, second order accurate and based on positive depth reconstruction. The CUDA parallelization led to speedups of two orders of magnitude with respect to a single-core CPU. The code, already validated against several severe benchmark tests, was here applied to two real-world cases. To capture all the main characteristics of the flow at different scales and to describe road and railway embankments without the introduction of 1D relationships, a high-resolution mesh size (2-5 meters) was adopted. Since the ratio between physical and computational time is always high, the numerical scheme herein presented can be embedded in real-time simulation tools, which can provide accurate and fast predictions useful also for flood management of occurring flood events.
