特征列求解的改进并行算法  

Improved Parallel Algorithm of Solving Characteristic sets

作　　者：李红波[1] 吴素萍[1] 刘锐[1] 杨军[1] 

机构地区：[1]宁夏大学数学与计算机学院,宁夏银川750021

出　　处：《计算机仿真》2012年第11期189-193,311,共6页Computer Simulation

基　　金：国家自然科学基金项目(60963004);国家星火项目(2011GA880001)

摘　　要：针对具有因式分解特性的多项式方程组的特征列并行算法,存在着可有效加速问题域较为狭窄的问题,即对不具备因式分解特性的普通问题的求解过程没有起到加速作用,为缩短计算时间,在Maple系统下采用编程模型,采用零点定理的通用并行策略,提出并行策略与特性并行策略进行结合,通过实例进行验证,实验结果表明,改进并行算法不仅可以更为高效的求解具备因式分解特性的问题,而且也可以适用于普通多项式方程组。In the study of parallel algorithm of characteristic sets, to improve the efficiency of polynomial equation systems that can be factorized, the parallel algorithm can only accelerate the computing process of a narrow scope of problems but fails at those common problems that cannot be factorized. In order to overcome this shortcoming, a com- bining parallel algorithm of a common parallel strategy based on Zero Theorems and the previous one was introduced in this paper. Then experiment was performed to support this idea by utilizing the Task Programming Model in Maple. The experimental results show that this algorithm can not only improve the computing efficiency of the problems char- acterizing factorization furthermore, but also work well with those common problems that cannot be factorized.

关 键 词：吴方法 并行化 特征列 多线程 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]