基于改进蚁群算法的算力灵活迁移优化算法  

A Flexible Migration Algorithm for Arithmetic Power Based on Improved Ant Colony Algorithm

作　　者：鲍兴川 刘世栋 张宁 BAO Xingchuan;LIU Shidong;ZHANG Ning(State Grid Smart Grid Research Institute Co.,Ltd.,Nanjing 210003,Jiangsu Province,China;Information&Telecommunication Branch,State Grid Corporation of China,Xicheng District,Beijing 100761,China)

机构地区：[1]国网智能电网研究院有限公司,江苏省南京市210003 [2]国家电网有限公司信息通信分公司,北京市西城区100761

出　　处：《电力信息与通信技术》2024年第3期1-8,共8页Electric Power Information and Communication Technology

基　　金：国家电网有限公司总部管理科技项目资助“面向新型电力系统跨域分层算力网络协同调度关键技术研究及应用”(5700-202318269A-1-1-ZN)。

摘　　要：针对现有云计算环境下国网数据中心资源调度存在的调度效率低、能源消耗高等问题,文章提出了一种基于改进蚁群算法的算力灵活迁移优化算法。首先构建国网云数据中心的算力迁移模型,对数据中心的资源调度能耗进行建模。然后通过引入细菌觅食算法改进基本蚁群算法的信息素初始化,并重新设计了启发函数和信息素挥发因子。仿真实验结果表明,与现有模型相比,文章的算法能够求出更优的算力资源调度方案,在减小任务完成时间的同时降低了国网数据中心36.6%的能耗。To solve the problems of low scheduling efficiency and high energy consumption in State Grid data center resource allocation in existing cloud computing environments,this paper proposes a flexible migration algorithm for arithmetic power based on improved ant colony algorithm.Firstly,a computational migration model for State Grid cloud data centers is constructed to model the energy consumption of resource scheduling in data centers.Subsequently,by integrating the bacterial foraging optimization algorithm,we enhanced the initial pheromone distribution in the fundamental ant colony optimization algorithm and reformulated both the heuristic function and pheromone evaporation factor.Simulation outcomes demonstrate that,compared with existing models,the algorithm proposed in this paper can find a more optimal scheduling solution for computing resources.This algorithm not only abbreviates the task execution duration but also culminates in a substantial 36.6%reduction in energy consumption in the State Grid data centers.

关 键 词：算力网络 数据中心 改进蚁群算法 细菌觅食算法 资源调度 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]