检索规则说明:AND代表“并且”;OR代表“或者”;NOT代表“不包含”;(注意必须大写,运算符两边需空一格)
检 索 范 例 :范例一: (K=图书馆学 OR K=情报学) AND A=范并思 范例二:J=计算机应用与软件 AND (U=C++ OR U=Basic) NOT M=Visual
作 者:喻高远 楼云锋[1,2,3] 李俊杰 金先龙 YU Gaoyuan;LOU Yunfeng;LI Junjie;JIN Xianlong(State Key Laboratory of Mechanical System and Vibration,Shanghai Jiao Tong University,Shanghai 200240,China;School of Mechanical Engineering,Shanghai Jiao Tong University,Shanghai 200240,China;Aerospace System Engineering Shanghai,Shanghai 201108,China)
机构地区:[1]上海交通大学机械系统与振动国家重点实验室,上海200240 [2]上海交通大学机械与动力工程学院,上海200240 [3]上海宇航系统工程研究所,上海201108
出 处:《振动与冲击》2023年第6期152-158,共7页Journal of Vibration and Shock
基 金:国家自然科学基金(11772192)。
摘 要:根据国产申威异构众核分布式存储计算机的体系结构特点,提出了一种结构瞬态有限元分层并行计算方法,对于提高国产申威异构众核分布式存储并行计算机下大型、超大型复杂结构系统的瞬态并行求解效率具有重要意义。该方法在分层通信和Newmark-HHT算法的基础上构建了大规模复杂结构系统的瞬态并行求解体系,不仅实现了计算过程中大量数据的分布式存储,显著改善了数据的内存访存效率;而且实现了计算过程的两层并行,有效改善了通信效率。因此,该计算方法能够充分利用国产申威异构众核分布式存储并行计算机的体系结构特点提升结构瞬态大规模并行计算效率。最后通过典型数值算例验证了该方法的正确性和有效性,并将其应用于某高层建筑,实现其上千万自由度、数万核的结构瞬态并行计算。According to the architecture characteristics of the domestic heterogeneous multi-core processor,a hierarchical communication parallel computing algorithm for structural transient analysis was proposed,which had important significance for improving the parallel efficiency of the system transient analysis on the entire large structure under the domestic heterogeneous multi-core and distributed memory parallel computers.Based on hierarchical communication and the Newmark-HHT algorithm,a parallel computing system for a large-scale transient analysis was established,which could not only significantly improve the memory access rate through the distributed storage of a large amount of data,but also significantly improve the communication rate with the two-layer parallelization of the computational procedure.It is shown that the method can improve the efficiency rates of parallel computing of the large-scale transient analysis by fully exploiting the architecture characteristics of the domestic heterogeneous multi-core and distributed memory parallel computers.Finally,typical numerical experiments were used to validate the correctness and efficiency of the proposed method.Then,the parallel transient analysis of a high-rise building with over ten-million-DOF was performed and ten thousands of core processors were applied.
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在链接到云南高校图书馆文献保障联盟下载...
云南高校图书馆联盟文献共享服务平台 版权所有©
您的IP:216.73.216.198