Research On Service Migration And Maintenance Methods In Mobile Edge Computing

The emergence of many interactive applications,such as Internet of Vehicles(IoV)and smart homes,has been made possible by the development and popularity of cloud services.However,the cloud computing paradigm is unable to meet the requirements of delay-sensitive applications.The proposal of mobile edge computing(MEC)has solved this problem.Mobile edge computing deploys part of services that originally located in cloud data centers at the edge of the network,bring computing and storage closer to users,thereby reducing communication delay and improving the quality of experience(QoE)for end users.However,there are some problems in the mobile edge computing environment:(1)Due to the mobility of users,the MEC servers that can provide services for users also change dynamically.Therefore,the edge services requested by the users have to be migrated between edge nodes.The user is forced to access the edge service running on the pervious serving edge node until the migration of the edge service is completed.However,forwarding the service request from the MEC server that the user is connecting to another MEC server will significantly increase edge service response time,and thus drastically reduce the user QoE.(2)With the rapid growth of the number of edge services,it becomes difficult to select optimal edge services that meet users' need from a lot of edge services that have the same functions.In addition,the performance of MEC is dynamic in nature due to its location awareness,mobility and proximity.As a result,an effective mechanism is needed for providing efficient dynamic service maintenance for edge services.To solve the above problems,this thesis first proposes an approach to reduce service migration based on transmission power control.Then,we also propose the Skyline Graph Model(SGM)and related algorithms to realize efficient dynamic service maintenance for edge services.The main tasks of this thesis include:(1)In order to reduce the QoE degradation caused by service migrates,this thesis proposes a method for reducing service migration based on transmission power control.First of all,the basic idea of Voronoi Diagram is used to divide the users' activity range into multiple sub-areas according to the transmission power of the MEC servers.Each sub-area contains a MEC server to serve users in the area.Then,adjust the size of each sub-area by controlling the transmission power of the MEC servers to achieve two goals: minimizing the number of users' wireless handovers,and the energy consumption of MEC servers.Then,particle swarm optimization(PSO)is used to solve this multi-objective optimization problem.Extensive experiments are conducted to demonstrate the effectiveness and efficiency of the algorithm.(2)In order to realize efficient dynamic service maintenance for edge services,this thesis proposes the use of the Skyline Graph Model which can provide the storage and update of edge services based on the theory of directed acyclic graphs.Given the dominance relationship and transitivity of the Skyline,it can achieve the efficient storage and dynamic update of edge services,and then solve the fast selection problem of edge services.Besides,Skyline Graph Algorithm is proposed based on the Skyline Graph Model,which can optimize the insertion,deletion,update and search of the edge services to realize efficient dynamic maintenance of edge services.At last,comprehensive experiments that are based on both real-world and simulated datasets are conducted and the results successfully demonstrate the effectiveness and robustness of Skyline Graph Algorithm.