作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2021, Vol. 47 ›› Issue (11): 37-43. doi: 10.19678/j.issn.1000-3428.0061632

• 热点与综述 • 上一篇    下一篇

基于时延与能量感知的边缘服务器放置方法

赵兴兵1, 赵一帆2, 李波1, 陈春3, 丁洪伟1   

  1. 1. 云南大学 信息学院, 昆明 650500;
    2. 云南民族大学 电气信息工程学院, 昆明 650504;
    3. 云南民族大学 应用技术学院, 昆明 650504
  • 收稿日期:2021-05-13 修回日期:2021-07-17 发布日期:2021-07-22
  • 作者简介:赵兴兵(1997-),男,硕士研究生,主研方向为移动边缘计算;赵一帆,博士研究生;李波,副教授、博士;陈春,讲师、硕士;丁洪伟,教授、博士。
  • 基金资助:
    国家自然科学基金(61461053);云南大学研究生科研创新项目(2020306)。

Edge Server Placement Method Based on Time Delay and Energy Awareness

ZHAO Xingbing1, ZHAO Yifan2, LI Bo1, CHEN Chun3, DING Hongwei1   

  1. 1. School of Information, Yunnan University, Kunming 650500, China;
    2. School of Electrical Information Engineering, Yunnan Minzu University, Kunming 650504, China;
    3. School of Applied Technology, Yunnan Minzu University, Kunming 650504, China
  • Received:2021-05-13 Revised:2021-07-17 Published:2021-07-22

摘要: 针对移动边缘计算中无线城域网环境下的边缘服务器放置(WESP)问题,建立时延和能耗模型并将WESP问题转化为带约束条件的单目标优化问题,进而提出一种基于混沌麻雀搜索算法的边缘服务器放置方法。使用精英反向学习策略初始化种群,增加初始种群的多样性,加快算法搜索速度。通过设计新的个体编码方式准确描述WESP问题,优化算法更新过程。采用逻辑混沌映射策略改进麻雀个体,保证迭代后期的种群多样性,加快算法收敛速度。仿真结果表明,与主流放置方法相比,该方法在时延和能耗优化方面表现突出,并且系统开销下降了18.1%。

关键词: 边缘计算, 移动边缘计算, 边缘服务器, 麻雀搜索算法, 时延, 能耗

Abstract: To address the Edge Server(ES) placement problem in the Wireless Metropolitan Area Network(WMAN) environment of mobile edge computing,the time delay and energy consumption of servers are modeled to simplify the WESP problem into a single-objective optimization problem with constraints.On this basis,an edge server placement method based on the Chaotic Sparrow Search Algorithm(CSSA) is proposed.The method uses an Elite Opposition-Based Learning(EOBL) method to initialize the population,which increases the diversity of the initial population and speeds up the search of the algorithm.Then a new individual coding method is proposed to effectively describe the WESP problem and improve the update process of the algorithm.The sparrow individuals are improved by using the logical chaotic mapping method to ensure the population diversity at the later stage of the iteration and accelerate the convergence of the algorithm.Simulation results indicate excellent performance of the proposed method in optimization time delay and energy consumption compared with the mainstream placement methods,and a reduction of 18.1% in system overhead.

Key words: edge computing, mobile edge computing, Edge Server(ES), Sparrow Search Algorithm(SSA), time delay, energy consumption

中图分类号: