| 基于FTM算法的GPU加速 |
| |
| 引用本文： | 曾良,杜煜昊,张莹,胡昱,洪瑶,陈虎.基于FTM算法的GPU加速[J].计算力学学报,2017,34(4):511-516. |
| |
| 作者姓名： | 曾良 杜煜昊 张莹 胡昱 洪瑶 陈虎 |
| |
| 作者单位： | 南昌大学 机电工程学院,南昌,330031 |
| |
| 基金项目： | 国家自然科学基金（11562011）；江西省自然科学基金（20151BAB202002）资助项目. |
| |
| 摘 要： | 为了解决FTM(Front Tracking Method)算法在计算机中计算耗时长的问题,利用CUDA(Compute Unified Device Architecture)来实现FTM算法在GPU中的并行计算。结合GPU并行计算架构的特性以及FTM算法的特点,本文通过共享内存的引入、线程块划分和线程块共享内存边界元素的纳入、迭代方法的改进和迭代过程中存储结构的变换等方法,提出了将FTM算法中的网格计算以及界面标记点处理方法在GPU中的实现方式。最后,通过模拟单气泡在静止液体中的自由上升运动,验证了FTM在GPU中计算的可行性与计算效率的提升。
|
| 关 键 词： | FTM CUDA GPU 并行计算 计算流体动力学 数值模拟 |
| 收稿时间： | 2016/4/5 0:00:00 |
| 修稿时间： | 2016/8/24 0:00:00 |
FTM algorithm based on GPU acceleration |
| |
| ZENG Liang,DU Yu-hao,ZHANG Ying,HU Yu,HONG Yao,CHENG Hu.FTM algorithm based on GPU acceleration[J].Chinese Journal of Computational Mechanics,2017,34(4):511-516. |
| |
| Authors: | ZENG Liang DU Yu-hao ZHANG Ying HU Yu HONG Yao CHENG Hu |
| |
| Institution: | School of Mechanical and Electrical Engineering, Nanchang University, Nanchang 330031, China,School of Mechanical and Electrical Engineering, Nanchang University, Nanchang 330031, China,School of Mechanical and Electrical Engineering, Nanchang University, Nanchang 330031, China,School of Mechanical and Electrical Engineering, Nanchang University, Nanchang 330031, China,School of Mechanical and Electrical Engineering, Nanchang University, Nanchang 330031, China and School of Mechanical and Electrical Engineering, Nanchang University, Nanchang 330031, China |
| |
| Abstract: | In order to solve the problem of low computational effective of the FTM algorithm,this paper used CUDA to realize the parallel implementation of the FTM algorithm in GPU.Combining with the GPU parallel computing architecture and the characteristics of the FTM algorithm,and through introducing shared memory and bringing thread partition and thread block shared memory boundary elements into use,we proposed a method to implement the grid implementation of the FTM algorithm and a way of processing interface marked point in the GPU.At last,the feasibility and efficiency of FTM in GPU were verified by simulating the free ascending motion of single bubbles in a stationary liquid. |
| |
| Keywords: | FTM CUDA GPU parallel computing computational fluid dynamics numerical simulation |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
| 点击此处可从《计算力学学报》浏览原始摘要信息 |
| 点击此处可从《计算力学学报》下载免费的PDF全文 |
|
