Parallel Optimization of CFD Core Algorithms Based on Domestic Processor

doi:10.11871/jfdc.issn.2096-742X.2021.04.008

Abstract

Abstract:

[Objective] In order to accelerate the calculation of domestic CFD software, this paper designs and implements an accelerated version of CFD core algorithms based on the domestic processor. [Methods] Based on the CCFD V3.0 version, the core algorithms of the software were ported to the domestic processor and optimized by various methods. [Results] Using a 128*128*128 grid for experiments, the simulation results of the ported program are basically the same as the original CPU version. Compared with a single CPU core, one acceleration card has achieved 166 times acceleration on convection calculations, and 59 times acceleration on ADI iterative calculations. [Limitations] Because there are many modules in CFD software, the target CFD program has not been ported and optimized entirely. In the future, the software will be ported completely to the domestic processors for acceleration. [Conclusions] This article has realized the parallel optimization of the CFD core algorithms based on domestic processors and achieved good acceleration results, which provides experience and reference for porting and optimizing CFD software in the future.

Key words: domestic processors, CFD, transplantation, high-performance computing, parallel computing, optimization

CAO Yikui,LU Zhonghua,ZHANG Jian,LIU Xiazhen,YUAN Wu,LIANG Shan. Parallel Optimization of CFD Core Algorithms Based on Domestic Processor[J]. Frontiers of Data and Computing, 2021, 3(4): 93-103.

Figures/Tables 13

Fig.1

Fig.2

Fig.3

Fig.4

Fig.5

Table 1

Fig.6

Fig.7

Table 2

Table 3

Table 4

Fig.8

Fig.9

References 22

[1]	张兵, 韩景龙. 基于GPU和隐式格式的CFD并行计算方法[J]. 航天学报, 2010,31(2):249-256.
[2]	OBRECHT C, KUZNIK F, TOURANCHEAU B, et al. A new approach to the lattice Boltzman method for grap-hics processing units[J]. Computers & Mathematics with Applications, 2010,61(12):3628-3638. doi: 10.1016/j.camwa.2010.01.054
[3]	CORRIGAN A, CAMELLI F F, LONHNER R, et al. Running unstructured grid-based CFD solvers on modern graphics hardware[J]. International Journal for Numerical Methods in Fluid, 2010,62:86-90.
[4]	JESPERSEN D C. Acceleration of a CFD code with a GPU[J]. Scientific Programming, 2010,18(3):193-201. doi: 10.1155/2010/564806
[5]	李大力, 张理论, 徐传福, 等. 雅可比迭代的CPU/GPU并行计算及在CFD中的应用[C/OL]. [2012-08-21]. http://www.ccf.org.cn/sites/ccf/weekly/zhuanti/gaoxing-nengjisuan/雅可比迭代的CPU/并行计算及在CFD中的应用.pdf .
[6]	董廷星, 李新亮, 李森, 迟学斌. GPU上计算流体力学的加速[J]. 计算机系统应用, 2011,20(01):104-109
[7]	邓亮, 徐传福, 刘巍, 张理论. 交替方向隐式CFD解法器的GPU并行计算及其优化[J]. 计算机应用, 2013,33(10):2783-2786.
[8]	Emelyanov, Vladislav N, A. G. Karpenko, K. N. Volkov. Development of advanced computational fluid dynamics tools and their application to simulation of internal turbu-lent flows[J]. Progress in Flight Physics-Volume 7, 2015,7(1):247-268.
[9]	Emelyanov, Vladislav, Anton Karpenko, Konstantin Volkov. Development and acceleration of unstructured mesh-based CFD solver[J]. Progress in Flight Physics-Volume 9, 2017,9(1):387-408.
[10]	Lai Jianqi, Hua Li, Zhengyu Tian. CPU/GPU Heterog-eneous Parallel CFD Solver and Optimizations[C]. In Proceedings of the 2018 International Conference on Service Robotics Technologies, ACM, 2018,3(1):88-92.
[11]	Lai Jianqi et al. A Multi-GPU Parallel Algorithm in Hypersonic Flow Computations[J]. Mathematical Pro-blems in Engineering, 2019,10(1):1-15.
[12]	党冠麟, 刘世伟, 胡晓东, 张鉴, 李新亮. 基于CPU/GPU异构系统架构的高超声速湍流直接数值模拟研究[J]. 数据与计算发展前沿, 2020,2(01):105-116.
[13]	张云泉, 袁良, 袁国兴, 李希代. 2019年中国高性能计算机发展现状分析与展望[J]. 数据与计算发展前沿, 2020,2(1):18-26.
[14]	陈刚, 王磊, 陆忠华, 梁贤. 万核级并行飞机气动模拟软件CCFD研制[J]. 华中科技大学学报(自然科学版), 2011,39(S1):99-101.
[15]	钱德沛. 构建支撑科技创新的新一代计算基础设施[J]. 数据与计算发展前沿, 2020,2(1):1-17.
[16]	Roe P L. Approximate Riemann solvers, parameter vectors and differences schemes[J]. Journal of Comput-ational Physics, 1981,43(2):357-372.
[17]	Van Leer B. Flux vector splitting for Euler equations[J]. Lecture Notes in Physics, 1982,170.
[18]	Liou M-S. Progress towards an improved CFD method: AUSM+[C]//AIAA Paper 95-1701, 1995.
[19]	Jameson A, Schmidt W, Turkel E. Numerical solutions of the Euler equations by finite volume methods using Runge-Kutta time-stepping[C]//AIAA Paper 81-1259, 1981.
[20]	阎超. 计算流体力学方法及应用[M]. 北京: 北京航空航天大学出版社, 2006: 154-155.
[21]	ZHANG L P, WANG Z J. A block LU-SGS implicit dual time-step ping algorithm for hybrid dynamic meshes[J]. Computer & Fluids, 2004,33(7):891-916. doi: 10.1016/j.compfluid.2003.10.004
[22]	John Cheng, Grossman Max, McKercher Ty. Professional CUDA C Programming. China Machine Press[M]. 2017: 19-35.