A parallel meshless dynamic cloud method on graphic processing units for unsteady compressible flows past moving boundaries

Abstract

This paper presents an effort to implement a recently proposed meshless dynamic cloud method [Hong Wang et al. A study of gridless method with dynamic clouds of points for solving unsteady CFD problems in aerodynamics, Int. J. Numer. Meth. Fluids 2010; 64: 98-118] on modern high-performance graphic processing units (GPUs) with the compute unified device architecture (CUDA) programming model. Within the framework of the meshless method, clouds of points used as basic computational stencils are distributed in the whole flow domain. The spatial derivatives of the governing equations are discretised by the moving-least square scheme on every cloud of points. Roe's approximate Riemann solver is adopted to compute the convective flux. A dual-time stepping approach, which iterates in physical and pseudo temporal spaces, is employed to obtain the time-accurate solution. Simulation of steady compressible flows over a fixed aerofoil is firstly carried out to verify the GPU implementation of the method. Then it is extended to compute unsteady flows past oscillatory aerofoils. Numerical outcomes are compared with experimental and/or other reference results to validate the method. Significant performance speedup of more than an order of magnitude is verified by the numerical results. Systematic analysis shows that GPU is more energy efficient than CPU for solving aerodynamic problems. This demonstrates the potential of the proposed method to solve fluid-structure interaction problems.

A GPU based compressible multiphase hydrocode for modelling violent hydrodynamic impact problems

“…Since 3D compressible multiphase flow simulations request very much more intensive computation than 1D problems, it is necessary to consider parallel computing implementations via many-core GPUs. The benefits of utilising GPU acceleration in CFD can be found in recent works pertaining to single-phase flows which successfully accelerated the flow solvers by several times and even orders of magnitude [37][38][39][40][41]. It is very likely that applying many core GPU techniques to dramatically reduce the computing time for simulating compressible multiphase flows will be attractive and beneficial to the academic and industrial communities.…”

A GPU-accelerated implicit meshless method for compressible flows

“…More colors will request more kernels to be launched, and more kernels will cause heavier overhead cost. Of course, this could also be influenced by the data locality issue [24]. These problems will be further investigated and addressed in our future work.…”

“…In meshless discretization [18][19][20][21][22][23][24] of the partial differential equations for CFD like Equation 1 [18], radius basis functions [19], conservative meshless schemes [20]. In the present work, a weighted least-square curve fit based meshless method [28] is applied and the spatial derivative coefficients can be obtained by solving the following linear system…”

“…Parallelization of these new methods on many-core graphics processors to calculate complex compressible flows more efficiently will undoubtedly be beneficial to scientific research and engineering applications. Recently some researchers have attempted to implement explicit meshless methods on GPUs to calculate 2D compressible flows [23,24]. However, it remains obscure whether implicit meshless methods, which converge much faster than explicit meshless methods on CPUs, would be able to be ported to GPUs to achieve further acceleration.…”

“…One of the biggest challenges in realizing implicit methods on the GPU is these methods' inherent data dependency characteristics, which will inevitably cause thread-racing conditions that could corrupt the data on the computer [24]. It is relatively easy to modify explicit algorithms to avoid thread-racing conditions, but it is much harder to achieve the same objective for implicit methods.…”