Research On Efficient Data Migration Service And Incentive Mechanism In Vehicle Edge Computing

In Vehicle Edge Computing(VEC),the execution of offloaded task needs not only the task data uploaded by the requesting vehicle,but also the additional data(such as database,link library and other task execution environment data)to support the task be executed successfully,and how to efficiently cache and access these supporting data becomes the key issue for task offloading.Since the VEC server that provides the task offloading service may not have all supporting data for task execution,the needed data should be obtained from other servers via data migration.This thesis focuses on the optimization of data migration delay in VEC and designs efficient data collaborative caching strategy and data migration incentive mechanism,considering the selfishness of data migration service parties.The main research contents are as follows:(1)A VEC architecture based on software-defined network is proposed,and the data migration method supporting data and the cache cooperation mechanism between VEC servers are analyzed.In order to minimize the migration delay of supporting data,an efficient data collaborative caching strategy based on service area dividing,server grouping and storage partition is proposed.This strategy divides the VEC service area into dense areas and ordinary areas according to the vehicle density,and according to the similarity between the requested data The server is divided into multiple groups,and the storage space of the server is divided into three parts,so as to take into account the data requirements of the VEC server itself,the group where it is located,and the global data,so as to achieve efficient utilization of storage resources.Experiments show that the proposed caching strategy has better performance in terms of average data migration latency compared with other strategies.(2)Considering the limited resources of edge servers,the idle resources of idle vehicles are aggregated and an vehicle resource aggregation platform is constructed.Considering the selfishness of each service provider,a data migration incentive mechanism based on edge-platform collaboration is designed.In this mechanism,the VEC server executing tasks can send data migration requests to vehicle resource aggregation platform and the VEC server with supporting data.Aiming at the interaction between the three participants,the utility function of each participant is reasonably defined,a two-stage Stackelberg game model is established,and the above-mentioned Stackelberg game equilibrium is proved to be unique by reverse induction analysis.In addition,a gradient optimization algorithm is proposed to quickly determine the optimal policy set among the three participants.Finally,through extensive simulation experiments,the effectiveness and superiority of the proposed data migration incentive mechanism are verified.