| With the development of the intelligent vehicular industry and applications,the requirements for vehicular sensing,computation,and communication are increasing.The vehicular edge computing architecture that integrates mobile edge computing and Internet of Vehicles,by deploying roadside units,edge computing servers,cloud servers,and other computing resources,sinks the network to vehicle terminals,which can effectively solve the problem of the insufficient computing capacity of vehicles and the demand for communication and mobility management.Virtualization is the supporting technology for edge computing,which can effectively distribute and manage tasks.Container technology,as the latest advancement in virtualization technology,relies on lightweight,isolated,and flexible management to greatly improve the availability of vehicle tasks.In the IoV environment,the available computing resources and communication network topology changes due to vehicle mobility lead to unavailability of computing services and affect user experience.Therefore,how to meet the requirements of application latency and service continuity by reasonably selecting the communication and computational resources in the scenario and performing computational migration is the focus of this paper.This paper firstly models and analyzes the migration process of serial containerized computing tasks in vehicular edge computing scenario of Cellular-Vehicle-to-Everything(C-V2X),and propose a mobility-aware dynamic migration algorithm to select computing resources using vehicular link reliability and resource matching,while taking into account the effect of the current subtask migration decision on subsequent subtasks.This paper transforms the resource mapping problem into a graph-theoretic approach to solve it,and optimize task completion time in the whole migration process.Experiments show that the proposed algorithm can effectively shorten the completion time of tasks,improve the success rate of task completion,and have good adaptability and stability under different environmental conditions.Then,this paper analyzes the computation migration of container task in the heterogeneous vehicle networking environment with multiple types of communication and computation resources and proposes an adaptive algorithm for joint communication handover and computational migration.The algorithm selects communication nodes and computational nodes based on vehicle and environment information,and describes the problem as an integer linear programming problem with the goal of trade-off End-to-End latency and communication-computation migration overhead and adapts the resources to dynamically changing scenarios by using the branch and board method.The experimental results show that the proposed algorithm can effectively reduce the application end-toend delay,reduce the service interruption time,adapt to the high-speed changes of heterogeneous vehicular networks,and meet the demand of low delay and continuity of user services. |
