《计算机应用》唯一官方网站 ›› 2020, Vol. 40 ›› Issue (2): 386-391.DOI: 10.11772/j.issn.1001-9081.2019081401

• 2019年全国开放式分布与并行计算学术年会(DPCS 2019)论文 • 上一篇    下一篇

多种任务调度混合的IB-LBM并行优化方法

刘智翔1, 刘慧超1, 黄冬梅1,2(), 周丽萍3, 苏诚4   

  1. 1.上海海洋大学 信息学院,上海 201306
    2.上海电力大学,上海 200090
    3.上海大学 计算机工程与科学学院,上海 200444
    4.国家海洋局东海分局 东海预报中心,上海 200136
  • 收稿日期:2019-07-31 修回日期:2019-08-30 接受日期:2019-09-19 发布日期:2019-09-29 出版日期:2020-02-10
  • 通讯作者: 黄冬梅
  • 作者简介:刘智翔(1986—),男,湖南邵东人,讲师,博士,CCF会员,主要研究方向:高性能计算、计算流体力学
    刘慧超(1996—),男,山东潍坊人,硕士研究生,主要研究方向:高性能计算
    周丽萍(1984—),女,江西临川人,实验师,硕士,主要研究方向:并行计算
    苏诚(1962—),男,上海人,教授,主要研究方向:空间信息、海洋测绘。
  • 基金资助:
    上海市青年科技英才扬帆计划资助项目(18YF1410100);国家自然科学基金资助项目(91630206);上海市地方高校能力建设项目(17050501900);上海高校青年教师培养资助计划项目(ZZSHOU18012);上海海洋大学博士科研启动基金资助项目(A2-2006-00-200338)

IB-LBM parallel optimization method mixed with multiple task scheduling modes

Zhixiang LIU1, Huichao LIU1, Dongmei HUANG1,2(), Liping ZHOU3, Cheng SU4   

  1. 1.College of Information Technology,Shanghai Ocean University,Shanghai 201306,China
    2.Shanghai University of Electric Power,Shanghai 200090,China
    3.School of Computer Engineering and Science,Shanghai University,Shanghai 200444,China
    4.East China Sea Forecast Center,East China Sea Branch of the State Oceanic Administration,Shanghai 200136,China
  • Received:2019-07-31 Revised:2019-08-30 Accepted:2019-09-19 Online:2019-09-29 Published:2020-02-10
  • Contact: Dongmei HUANG
  • About author:LIU Zhixiang, born in 1986, Ph. D., lecturer. His research interests include high performance computing, computational fluid dynamics.
    LIU Huichao, born in 1996, M. S. candidate. His research interests include high performance computing.
    ZHOU Liping, born in 1984, M. S., experimentalist. Her research interests include parallel computing.
    SU Cheng, born in 1962, professor. His research interests include spatial information, hydrographic surveying and charting.
  • Supported by:
    the Shanghai Sailing Program(18YF1410100);the National Natural Science Foundation of China(91630206);the Program for the Capacity Development of Shanghai Local Colleges(17050501900);the Funding Program of Shanghai University Young Teacher Training(ZZSHOU18012);the Doctoral Scientific Research Start Foundation of Shanghai Ocean University(A2-2006-00-200338)

摘要:

在使用浸入边界-格子玻尔兹曼方法(IB-LBM)求解流场时,为了得出比较精确的结果,往往需要规模较大、较密集的流场网格,这就会造成模拟过程时间长的问题。为了提高模拟的效率,利用IB-LBM局部计算的特点,结合OpenMP中三种不同的任务调度方式,给出了IB-LBM的并行优化方法。在并行优化中混合使用三种任务调度方式,以弥补单一任务调度造成的负载不均衡问题;将IB-LBM进行结构化分解,测试每一结构部分的最优调度方式,根据实验结果选择最优的调度组合方式,而在不同线程数下,最优的组合方式是不同的。优化结果通过并行加速比来检验,可以得出:在线程数较少的情况下,加速比趋近于理想状态;在线程数较多的情况下,虽然线程开辟和销毁的额外时间消耗对性能的优化产生了影响,模型的并行性能仍有了很大的提升。流场的模拟结果显示,在进行并行优化后, IB-LBM对流固耦合问题模拟的准确性并没有受到影响。

关键词: 浸入边界法, 格子玻尔兹曼法, 流场, OpenMP, 任务调度, 并行, 加速比

Abstract:

When using Immersed Boundary-Lattice Boltzmann Method (IB-LBM) to solve the flow field, in order to obtain more accurate results, a larger and denser flow field grid is often required, which results in a long time of simulation process. In order to improve the efficiency of the simulation, according to the characteristics of IB-LBM local calculation, combined with three different task scheduling methods in OpenMP, a parallel optimization method of IB-LBM was proposed. In the parallel optimization, three task scheduling modes were mixed to solve the load imbalance problem caused by single task scheduling. The structural decomposition was performed on IB-LBM, and the optimal scheduling mode of each structure part was tested. Based on the experimental results, the optimal scheduling combination mode was selected. At the same time, it could be concluded that the optimal combination is different under different thread counts. The optimization results were verified by speedup, and it could be concluded that when the number of threads is small, the speedup approaches the ideal state; when the number of threads is large, although the additional time consumption of developing and destroying threads affects the optimization of performance, the parallel performance of the model is still greatly improved. The flow field simulation results show that the accuracy of IB-LBM simulation of fluid-solid coupling problems is not affected after parallel optimization.

Key words: immersed boundary method, lattice Boltzmann method, flow field, OpenMP, task scheduling, parallel, speedup

中图分类号: