针对基于PVM的桌面PC机联网而成的网络并行计算环境中,处理机的运算速度较快而处理机间的通信相对较慢,以及微机的内存有限的实际情况,从实用的角度出发,给出了基于PVM的网上求解有限元方程组的并行m-Step Jacob i PCG方法,该算法的矩...针对基于PVM的桌面PC机联网而成的网络并行计算环境中,处理机的运算速度较快而处理机间的通信相对较慢,以及微机的内存有限的实际情况,从实用的角度出发,给出了基于PVM的网上求解有限元方程组的并行m-Step Jacob i PCG方法,该算法的矩阵和向量采用行元素相邻单元贡献法实现有限元总体刚度矩阵和荷载向量的并行计算与组装,分块储存在各处理机上,其处理机间通信较少。并在1-4台桌面PC机连接成的局域网,PVM3.4 on W indow2000,VC 6.0并行计算平台上编程对该算法进行了数值试验,得到了较理想的结果。展开更多
基于EBE-PCG(element by element-preconditioned conjugate gradient)策略的并行算法不用形成总体刚度矩阵,而且无需进行三维模型的区域分解,从而提高了并行计算的速度和效率,是实现协同优化设计的性能函数快速分析技术的有效途径。文...基于EBE-PCG(element by element-preconditioned conjugate gradient)策略的并行算法不用形成总体刚度矩阵,而且无需进行三维模型的区域分解,从而提高了并行计算的速度和效率,是实现协同优化设计的性能函数快速分析技术的有效途径。文中详细介绍有限元EBE(element by element)的运算方法,给出EBE-PCG并行算法的实现步骤,最后在网络集群环境下,综合运用多种编程语言和分析工具,实现基于EBE-PCG策略的三维有限元并行计算。计算结果表明,该并行算法的计算误差小,并行效率高,适合于性能函数的快速求解。展开更多
The strategies that minimize the overall solution time of multiple linear systems in 3D finite element method (FEM) modeling of direct current (DC) resistivity were discussed. A global stiff matrix is assembled and st...The strategies that minimize the overall solution time of multiple linear systems in 3D finite element method (FEM) modeling of direct current (DC) resistivity were discussed. A global stiff matrix is assembled and stored in two parts separately. One part is associated with the volume integral and the other is associated with the subsurface boundary integral. The equivalent multiple linear systems with closer right-hand sides than the original systems were constructed. A recycling Krylov subspace technique was employed to solve the multiple linear systems. The solution of the seed system was used as an initial guess for the subsequent systems. The results of two numerical experiments show that the improved algorithm reduces the iterations and CPU time by almost 50%, compared with the classical preconditioned conjugate gradient method.展开更多
文摘针对基于PVM的桌面PC机联网而成的网络并行计算环境中,处理机的运算速度较快而处理机间的通信相对较慢,以及微机的内存有限的实际情况,从实用的角度出发,给出了基于PVM的网上求解有限元方程组的并行m-Step Jacob i PCG方法,该算法的矩阵和向量采用行元素相邻单元贡献法实现有限元总体刚度矩阵和荷载向量的并行计算与组装,分块储存在各处理机上,其处理机间通信较少。并在1-4台桌面PC机连接成的局域网,PVM3.4 on W indow2000,VC 6.0并行计算平台上编程对该算法进行了数值试验,得到了较理想的结果。
文摘基于EBE-PCG(element by element-preconditioned conjugate gradient)策略的并行算法不用形成总体刚度矩阵,而且无需进行三维模型的区域分解,从而提高了并行计算的速度和效率,是实现协同优化设计的性能函数快速分析技术的有效途径。文中详细介绍有限元EBE(element by element)的运算方法,给出EBE-PCG并行算法的实现步骤,最后在网络集群环境下,综合运用多种编程语言和分析工具,实现基于EBE-PCG策略的三维有限元并行计算。计算结果表明,该并行算法的计算误差小,并行效率高,适合于性能函数的快速求解。
基金Projects(40974077,41164004)supported by the National Natural Science Foundation of ChinaProject(2007AA06Z134)supported by the National High Technology Research and Development Program of China+2 种基金Projects(2011GXNSFA018003,0832263)supported by the Natural Science Foundation of Guangxi Province,ChinaProject supported by Program for Excellent Talents in Guangxi Higher Education Institution,ChinaProject supported by the Foundation of Guilin University of Technology,China
文摘The strategies that minimize the overall solution time of multiple linear systems in 3D finite element method (FEM) modeling of direct current (DC) resistivity were discussed. A global stiff matrix is assembled and stored in two parts separately. One part is associated with the volume integral and the other is associated with the subsurface boundary integral. The equivalent multiple linear systems with closer right-hand sides than the original systems were constructed. A recycling Krylov subspace technique was employed to solve the multiple linear systems. The solution of the seed system was used as an initial guess for the subsequent systems. The results of two numerical experiments show that the improved algorithm reduces the iterations and CPU time by almost 50%, compared with the classical preconditioned conjugate gradient method.
