基于节点的局部网格生成并行算法

结合基于节点的局部网格生成算法和约束D elaunay路径(CDP)划分区域技术,提出了基于节点的局部网格生成并行算法。使用CDP划分任意给定节点的求解域,能够消除局部网格生成过程中并行处理器间频繁的通讯量,并最大限度地降低了总刚度矩阵生成过程中处理器间的通讯量。文中的并行算法自然实现了网格生成和刚度矩阵计算的无缝连接,从而使基于节点的局部有限元方法的潜在并行效率得到充分展现。数值试验结果表明,该局部网格生成并行算法是有效的、可靠的,尤其当计算大规模的问题时,选取合适的处理器数,并行效率几乎都能达到90%。

A New and Efficient Node-Based Local Mesh Generation Parallel Algorithm

Aim.Since 2004,we have engaged in node-based local mesh generation (NLMG) research~().For NLMG,existing parallel algorithms are not only unsuitable but also poor in efficiency.So we propose a new and what we believe to be an efficient NLMG parallel algorithm.In the full paper,we explain a new parallel algorithm in detail;in this abstract,we just add some pertinent remarks to listing the three topics of our explanation:(A) CDP(Constrained Delaunay Path)-based domain partition algorithm;(B) NLMG parallel algorithm;(C) further discussion of NLMG parallel algorithm;under topic A,using the CDP to partition an arbitrary node-given domain,we can eliminate the frequent communications among processors in the local mesh generation process and decrease the communications to the fullest degree during the stiffness matrix computing;under topic B,we point out that the parallel algorithm naturally implements the seamless link between mesh generation and stiffness matrix computing,which fully demonstrates the potential efficiency of the node-based local finite element method;under topic C,we discuss the great effect of load balance on parallel efficiency and discuss also the appropriate numbers of subdomains and processors.Numerical experimental results,given in Tables 1 and 2 and Figs.5 through 7 in the full paper, show preliminarily that:(1) the NLMG parallel algorithm is effective and reliable,especially for large-scale problems;(2) if the number of chosen processors is appropriate,the parallel efficiency can reach nearly 90%.