基于PSO递阶进化的网络资源活跃度增强算法

发布时间：2018-03-09 14:00

本文选题：云计算　切入点：粒子群　出处：《科技通报》2015年10期 　论文类型：期刊论文

【摘要】：为了提高云计算环境下网络资源访问和调度能力,需要增强网络资源的活跃度,传统方法采用源信息系统最小方差粒子群优化算法实现资源活跃度增强调度,直接交互式多源信息的缺陷,导致信息访问的滞后和时延。提出一种基于粒子群(PSO)递阶进化的多出口网络资源活跃度增强算法,构建多出口网络资源调度和网络系统结构,粒子群进化按照属性的数据波动进行递阶分层,得到一个资源数据聚类的高密度区域,使得每一个初始种群中的个体都应有一个解,在多波束搜索PSO空间中实现粒子群PSO递阶进化,提高网络资源访问的活跃度。仿真实验表明,采用该算法,能避免粒子群在进行网络资源搜索调度过程中陷入局部最优,有效提高控制搜索精度,运行时间较短,能有效增强多出口网络资源的活跃度,进而提高了资源搜索成功率。
[Abstract]:In order to improve the ability of network resource access and scheduling in cloud computing environment, it is necessary to enhance the activity of network resources. The traditional method uses the minimum variance particle swarm optimization algorithm of source information system to realize the enhanced scheduling of resource activity. The defects of direct interactive multi-source information lead to the delay and delay of information access. A multi-exit network resource activity enhancement algorithm based on particle swarm optimization (PSO) hierarchical evolution is proposed to construct multi-outlet network resource scheduling and network system structure. In the evolution of particle swarm, a high density region of resource data clustering is obtained by hierarchical stratification according to the data fluctuation of attributes, so that each individual in the initial population should have a solution. Particle swarm PSO hierarchical evolution is realized in multi-beam search PSO space, and the activity of network resource access is improved. The simulation results show that the proposed algorithm can avoid PSO falling into local optimum in the process of searching and scheduling network resources. The control search accuracy is improved effectively and the running time is short. It can effectively enhance the activity of multi-outlet network resources and thus improve the success rate of resource search.
【作者单位】：安阳工学院计算机科学与信息工程学院;
【分类号】：TP393.01

【参考文献】