Research On Service Migration In 5G Network Mobile Edge Computing In Multiple Scenarios

With the development of wireless communication technology and the popularization of Internet of Things technology,the number of terminal devices and data traffic ushered in explosive growth,and the computing demand of mobile users is also increasing.In the traditional cloud computing architecture,users offload tasks to the cloud for processing.However,the cloud computing data center is far away from users,which fails to meet users’ requirements on data transmission rate and service quality.As a key technology to promote the development of 5G,mobile edge computing provides users with adjacent computing services by deploying resources to the edge of the network,which can effectively reduce the response delay of user requests.Due to the limited resources of the edge server and the movement of users,services need to be continuously migrated to the edge server with more remaining resources or closer to users to maintain low response delay.At present,there have been many researches on the service migration of user-exclusive service,while the research on the service migration of multiple users who share services has just started,so it is of great significance to research it.This thesis studies service migration under different scenarios in 5G network.Service migration scenarios are divided into 5G network macro base station scenario and 5G network small base station scenario according to the types of base stations where edge servers are located.Firstly,the service migration in the 5G network macro base station scenario is researched.In this scenario,multiple users who share services are far away from the edge servers that cause the response delay to rise.Therefore,the service migration problem is summarized as a multi-objective optimization problem to minimize the average response time delay and energy consumption of service migration.A service migration algorithm based on user mobility awareness is proposed to solve this problem.Experimental results show that the proposed algorithm can obtain lower average response delay and service migration energy consumption than other algorithms in 5G network macro base station scenario,and show greater advantages with the increase of the number of users.Then,the service migration in the 5G network small base station scenario is researched.In this scenario,a large number of users share the services,leading to excessive load on some edge servers and increased response delay.Therefore,the service migration problem is summarized as a multi-objective optimization problem to maximize the total optimization time of migration and total load gain of migration.A service migration algorithm based on edge server load awareness is proposed to solve the problem.Experimental results show that the proposed algorithm can obtain the average response delay close to the optimal algorithm and less service migration times than other algorithms in the5 G network small base station scenario.When the number of users or service deployment increases,the average response delay is lower than the optimal algorithm.Finally,this thesis designs and implements a service migration management system,and the system test results show that the system can select the corresponding service migration algorithm to complete the service migration in different scenarios.