基于成本感知的边缘服务器部署方法

Edge server deployment method based on cost awareness

针对移动边缘计算(mobile edge computing,MEC)中边缘服务器(edge server,ES)供应商的成本预算问题,建立一种以最小化时延和部署成本为目标的数学模型。通过归一化方法将其转化为单目标优化问题,提出一种基于交叉算法的鲸鱼优化算法的边缘服务器部署方法;采用精英反向学习策略构造新种群,提高种群的多样性和全局收敛速度;采用改进的非线性收敛因子平衡算法的整体开发能力和局部探索能力;利用纵横交叉策略提高算法跳出局部最优的能力。使用上海电信基站的真实数据集进行仿真,其结果表明,与其它4种算法相比,该算法在时延和部署成本方面的表现均优于其它算法,系统成本下降了42.1%。

Aiming at the cost budget problem of edge server(ES)suppliers in mobile edge computing(MEC),a mathematical model was established to minimize the delay and deployment cost,and it was transformed into a single objective optimization problem using normalization method.An edge server deployment method based on cross algorithm whale optimization algorithm was proposed.The elite reverse learning strategy was used to construct a new population and improve the diversity and global convergence rate of the population.The overall development ability and local exploration ability of the improved nonlinear convergence factor balance algorithm were adopted.The crossbar strategy was used to improve the ability of the algorithm to jump out of local optimal.The real data set of Shanghai Telecom base station was used for simulation.The results show that compared with other four algorithms,the proposed algorithm outperforms other algorithms in terms of delay and deployment cost,and the system cost is reduced by 42.1%.