龙格库塔间断有限元方法求解二维欧拉方程多GPU加速实现.pdfVIP

International Journal of Fluid Dynamics 流体动力学, 2018, 6(2), 15-22 Published Online June 2018 in Hans. /journal/ijfd /10.12677/ijfd.2018.62003 Accelerating the Runge-Kutta Discontinuous Galerking Method for Solving Two-Dimensional Flow on Multi GPUs Xingyu Zhou, Tiegang Liu LMIB and School of Mathematics and Systems Science, Beijing University of Aeronautics and Astronautics, Beijing th st th Received: May 15 , 2018; accepted: May 31 , 2018; published: Jun. 7 , 2018 Abstract It is time-consuming to use Runge-Kutta discontinuous Galerkin method to solver flow field. In this article, we use multi GPUs to accelerate computing of two-dimension NACA0012 airfoil. The flow mesh is divided into several blocks according to the number of GPUs. We specialize kernels with a one-element-per-thread strategy, and use MPI to communicate data among computing nodes. In the process of communication, we use CUDA stream and MPI non-blocking operation to overlap computation and communication. The result shows when compared with the serial CPU program, the speedup ratio of GPU code running on one, two, four GPUs is around 33, 59 and 108. Keywords RKDG, Multi GPUs, MPI 龙格库塔间断有限元方法求解二维欧拉方程的 多GPU加速实现 周星宇，刘铁钢 北京航空航天大学数学与系统科学学院，数学、信息与行为教育部重点实验室，北京 收稿日期：2018年5月15 日；录用日期：2018年5月31 日；发布日期：2018年6月7 日 摘 要 为解决龙格库塔间断有限元方法(RKDG)求解流场耗时的问题，本文应用二维NACA0012翼型作为测试算 文章引用: 周星宇, 刘铁钢. 龙格库塔间断有限元方法求解二维欧拉方程的多 GPU 加速实现[J]. 流体动力学, 2018, 6(2): 15-22. DOI: 10.12677/ijfd.2018.62003 周星宇，刘铁钢 例，使用多GPU加速求解。将流程网格按照GPU个数进行剖分，每个GPU计算一个网格区域。各计算节 点上设置核函数的线程数等于流场网格数，节点间的数据通信使用MPI (Message Passing Interface) 。 通信过程中采用CUDA流和MPI非阻塞操作以覆盖数据的传输和计算，减少通信代价。结果表明，与CPU 串行