基于遗传算法和虚拟化技术的网络配置方法  

A Network Configuration Method Based on Genetic Algorithm and Virtualization Technology

作　　者：孙美卫[1] SUN Mei-wei(Department of Information Technology, Quanzhou Vocational College of Economics and Business, Quanzhou Fujian 362000, China)

机构地区：[1]泉州经贸职业技术学院信息技术系,福建泉州362000

出　　处：《菏泽学院学报》2020年第2期15-21,共7页Journal of Heze University

基　　金：福建省中青年教师教育科研项目(jat171059)。

摘　　要：为实现大数据和计算机资源的优化配置,在分布式并行计算环境下,充分应用了光网络虚拟化和遗传算法,提出一种新的网络配置方法.首先,通过遗传算法推导光波长路径的最佳配置,以及分配给每个用户的计算机资源量;然后,基于用户和供应商之间在网络距离方面的公平性,提出了一种分割分配方案.该方案用于确定向每个用户提供计算机资源的供应商,以及所提供的计算机资源量;最后,生成或更新用户的分配候选供应商列表,选定候选供应商列表排名最前的供应商,并将其空闲计算机资源分配给用户,以完成网络配置.实验结果验证了所提方法的有效性,实现了大数据和计算机资源之间的网络距离公平性,以及计算机资源提供方之间的有效负载平衡.与其他同类方法相比,多项评估指标更优.In order to realize the optimal allocation of big data and computer resources,a new network configuration method is proposed in the distributed parallel computing environment with the full application of optical network virtualization and genetic algorithm.First of all,genetic algorithm is used to derive the optimal configuration of optical wavelength path and the amount of computer resources allocated to each user.Then,a partition allocation scheme is proposed based on the fairness of network distance between users and suppliers.This scheme is used to determine the suppliers and the provided amount of computer resources.Finally,it generates or updates the list of candidate suppliers for user allocation,selects the top suppliers,and allocates their idle computer resources to users to complete network configuration.The experimental results show that the proposed method is effective and has achieved the fairness of network distance between big data and computer resources,as well as the effective load balance between computer resource providers.This method has better evaluation indexes compared with other similar ones.

关 键 词：网络配置 大数据 遗传算法 网络虚拟化 计算机资源 

分 类 号：TP393.08[自动化与计算机技术—计算机应用技术]