Towards an Efficient Implementation of an Accurate SPH Method
A modified version of the Smoothed Particle Hydrodynamics (SPH) method is considered in order to overcome the loss of accuracy of the standard formulation. The summation of Gaussian kernel functions is employed, using the Improved Fast Gauss Transform (IFGT) to reduce the computational cost, while tuning the desired accuracy in the SPH method. This technique, coupled with an algorithmic design for exploiting the performance of Graphics Processing Units (GPUs), makes the method promising, as shown by numerical experiments.
A CUDA-based implementation of an improved SPH method on GPU
We present a CUDA-based parallel implementation on GPU architecture of a modified version of the Smoothed Particle Hydrodynamics (SPH) method. This modified formulation exploits a strategy based on the Taylor series expansion, which simultaneously improves the approximation of a function and its derivatives with respect to the standard formulation. The improvement in accuracy comes at the cost of an additional computational effort. The computational demand becomes increasingly crucial as problem size increases but can be addressed by employing fast summations in a parallel computational scheme. The experimental analysis showed that our parallel implementation significantly reduces the runti…
Advanced numerical treatment of an accurate SPH method
The summation of Gaussian kernel functions is an expensive operation frequently encountered in scientific simulation algorithms and several methods have been already proposed to reduce its computational cost. In this work, the Improved Fast Gauss Transform (IFGT) [1] is properly applied to the Smoothed Particle Hydrodynamics (SPH) method [2] in order to speed up its efficiency. A modified version of the SPH method is considered in order to overcome the loss of accuracy of the standard formulation [3]. A suitable use of the IFGT allows us to reduce the computational effort while tuning the desired accuracy into the SPH framework. This technique, coupled with an algorithmic design for exploit…