A CFD Heterogeneous Parallel Solver Based on Collaborating CPU and GPU

3rd International Conference on Mechanical and Aeronautical Engineering

机械制造;航空航天工程

Lai, Jianqi^1 ; Tian, Zhengyu^1 ; Li, Hua^1 ; Pan, Sha^1

College of Aerospace Science and Engineering, National University of Defense Technology, Deya Road Kaifu District, Changsha, Hunan, China^1

关键词: Computational accuracy;    Computational precision;    Compute Unified Device Architecture(CUDA);    Floating-point computation;    Graphic processing unit(GPU);    Numerical results;    Parallel efficiency;    Parallel solutions;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/326/1/012012/pdf
DOI  :  10.1088/1757-899X/326/1/012012

学科分类：航空航天科学

来源: IOP

【 摘 要 】

Since Graphic Processing Unit (GPU) has a strong ability of floating-point computation and memory bandwidth for data parallelism, it has been widely used in the areas of common computing such as molecular dynamics (MD), computational fluid dynamics (CFD) and so on. The emergence of compute unified device architecture (CUDA), which reduces the complexity of compiling program, brings the great opportunities to CFD. There are three different modes for parallel solution of NS equations: parallel solver based on CPU, parallel solver based on GPU and heterogeneous parallel solver based on collaborating CPU and GPU. As we can see, GPUs are relatively rich in compute capacity but poor in memory capacity and the CPUs do the opposite. We need to make full use of the GPUs and CPUs, so a CFD heterogeneous parallel solver based on collaborating CPU and GPU has been established. Three cases are presented to analyse the solver's computational accuracy and heterogeneous parallel efficiency. The numerical results agree well with experiment results, which demonstrate that the heterogeneous parallel solver has high computational precision. The speedup on a single GPU is more than 40 for laminar flow, it decreases for turbulent flow, but it still can reach more than 20. What's more, the speedup increases as the grid size becomes larger.