吉林大学学报(理学版) ›› 2019, Vol. 57 ›› Issue (3): 647-652.

• 计算机科学 • 上一篇    下一篇

基于集合理论的求解Ramsey数算法

白云霄   

  1. 陕西科技大学 文理学院, 西安 710021
  • 收稿日期:2018-03-03 出版日期:2019-05-26 发布日期:2019-05-20
  • 通讯作者: 白云霄 E-mail:baiyunxiao@sust.edu.cn

Solving Ramsey Number Algorithm Based on Set Theory

BAI Yunxiao   

  1. School of Arts and Sciences, Shaanxi University of Science & Technology, Xi’an 710021, China
  • Received:2018-03-03 Online:2019-05-26 Published:2019-05-20
  • Contact: BAI Yunxiao E-mail:baiyunxiao@sust.edu.cn

摘要: 针对传统单核DNA计算机算法求解Ramsey数时运算效率较低, 求解过程耗时高, 所得结果误差较大的问题, 提出一种基于集合理论的求解Ramsey数算法. 该算法以基于集合理论的MapReduce模型中Phoenix++系统为基础, 设计单核CPU下的圈集对完全图的Ramsey数求解算法并对其实施优化, 优化时进行数据预处理、 高效任务分割和键值对规划等过程, 获取根据Phoenix++系统基于集合理论的并行算法, 采用DNA计算机算法求解Ramsey数, 并对其数值进行验证, 实现Ramsey数的求解. 实验结果表明, 程序处理图像数量随着顶点数的增加而不断增大, 该方法求解Ramsey数的正确性较高, 最大加速比和执行效率较好, 运算性能较强.

关键词: 集合理论, 计算机算法, Ramsey数求解, MapReduc模型, 单核CPU

Abstract: Aiming at the problem of low efficiency, high timeconsuming and large error in solving Ramsey number by the traditional singlecore DNA computer algorithm, the author proposed a Ramsey number algorithm based on set theory. The algorithm was based on the Phoenix++ system of the MapReduce model based on set theory. The author designed and optimized the Ramsey number algorithm for the complete graph under the singlcore CPU. Data preprocessing, efficient task segmentation and keyvalue pair planning were performed during optimization. The parallel algorithm based on set theory in Phoenix++ system was obtained. The Ramsey number was solved by DNA computer algorithm and its numerical value was verified to solve the Ramsey number. The experimental results show that the number of image processed by the program increases with the increase of the vertices. The proposed method has high accuracy in solving the Ramsey number, better maximum acceleration ratio and execution efficiency, and strong computational performance.

Key words: set theory, computer algorithm, Ramsey number solution,  , MapReduc model, singlecore CPU

中图分类号: 

  • TP301.6