Research On Cloud System Resource Allocation Strategy Based On Vehicle Network Application

With the rapid development of the global automotive industry,many automotive applications have emerged.These applications require powerful computing resources,sufficient bandwidth resources and a large amount of storage space.However,the space of a single vehicle is small.Meanwhile,it is required to meet low-cost requirements.Only relying on the equipment in the vehicle is no longer sufficient to provide resources for a variety of applications.Cloud computing provides users with flexible and scalable services and enables applications to access computing power,storage space,and information services on demand.Therefore,it is of great significance to study the resource allocation of Internet of Vehicles in the cloud system.The main contributions of this thesis are as follows.(1)To solve the problem of request interruption and high blocking rate due to vehicle mobility,this thesis proposes a resource allocation strategy for vehicle networking cloud system based on multi-objective evolutionary algorithm.With the goal of minimizing blocking rate and cost,a multi-objective resource allocation model is established.Nondominated sorting genetic algorithm II(NSGA-II))is used to solve the proposed algorithm.In the population initialization phase,a matching factor is defined to assess the providing resources ability of vehicles.In the cross and mutation stages,dynamic crossover probability and mutation probability are set according to the individual’s fitness in the population,which increases the diversity of the population and avoids local optimization.Simulation results show that the proposed algorithm can effectively reduce the blocking rate and cost.(2)To solve the problem of different requirements of different types of applications and request migration,this thesis proposes a migration-aware resource allocation strategy for vehicle networking cloud systems.Vehicle resources are divided into two parts: reliable resources and flexible resources.The reliable resources are used to provide resources for high priority requests and low priority requests which are coming at the current time.The flexible resources are used to provide resources for requests that need to be migrated.In order to reserve resources for requests reasonably,the size of reliable resources and flexible resources at the current moment are calculated by the number of migrated requests at the previous moment,the weight of flexible resources at the previous moment and the resources of high-priority requests of the migration.The reservation mechanism can reduce the situation in which the request is blocked due to the departure of the vehicles.To prioritize the need for high priority requests,the objective function is set to minimize the blocking rate,minimize the interruption rate of the migration request and maximize the proportion of high priority requests.The simulation results show that the proposed algorithm can effectively reduce the blocking rate,reduce the interrupt rate of migration requests,and meet more high-priority requests.