Research On Task Clustering And Scheduling Simulation Of IoV System

The vigorous development of the Internet of Things and communication technologies has created opportunities for the rise of the Internet of Vehicles.The vehicle collects a large amount of data information about itself and the surrounding environment through various sensing devices,and then provides users with various Internet of vehicle services through data processing.However,with the explosive growth of networked vehicles,it is difficult for a single on-board unit to meet computing storage and bandwidth requirements.Therefore,the rise of cloud computing technology and MEC technology has injected new impetus into the development of Internet of Vehicles.The researchers achieved the goal of breaking through resource constraints by scheduling vehicle tasks to cloud computing platforms or MEC servers.However,the traditional Internet of Vehicles task scheduling strategy merely schedules tasks that need to be scheduled by the vehicle-mounted terminal to the MEC server or the core cloud data center,and does not consider scheduling among the three,resulting in waste of resources.Therefore,this paper has conducted research on this issue:1.This paper studies the clustering problem of vehicle-mounted tasks in the IoV system.Through clustering,vehicle tasks are divided into computing demand type,band pass demand type and storage demand type.Then this paper evaluates the computing unit to get the resource tags that the computing unit mainly owns and prepares for the evaluation of the fitness value of the next task scheduling.2.This paper studies the scheduling strategy for clustered tasks.For the task scheduling strategy in the IoV system,this paper uses genetic algorithm.In the genetic algorithm,this paper combines the greedy strategy and the method of reorganization and intersection to redesign the intersection operator.And the task scheduling strategy evaluation is designed.3.This paper analyzed and compared three common cloud computing simulation platforms,and then selected CloudSim to simulate the task scheduling strategy.Simulation results show that the task scheduling strategy can effectively improve the comprehensive optimization of resource matching and execution time,while reducing the total execution time of the task and improving the resource utilization of the computing unit.