Research And Demonstration Of Task Assignment Strategy Based On MEC

Since the advent of cloud computing in 2005,the way we live,study and work has changed dramatically.The explosive growth of data is the main theme in the era of big data.With the deepening of the Internet of everything,the growth rate of data far exceeds network bandwidth.At the same time,the emergence of many new applications,such as intelligent manufacturing and unmanned driving,has put forward higher requirements for delay,and the traditional cloud computing framework has been difficult to support the Internet of everything applications in real time and efficiently.It is one of the important methods to reduce the response delay and the data transmission delay to deal with the task nearby.This is how MEC(Mobile Edge Computing)handles the requests.Therefore,for the mass data generated by the terminal and its real-time requirements,how to make full use of the computing resources of the server to process these data and feedback the processing results in real time is an urgent problem to be solved.In view of this,this paper designs an Internet of things platform based on MEC server,which takes the granularity of the request task as the starting point of the problem.Based on the computer network,the problem of task unloading is modeled and the mixed integer nonlinear programming problem of task unloading path is formed,and then selects the optimization scheme according to the actual background of the problem.After the task scheduling of a single server is completed,the load balancing among the servers is completed through the detection of the surrounding servers,so as to obtain the optimal processing process of the whole task from the terminal to the system.The experimental results show that the system is better than the cloud computing model in reducing response delay.The main research of this paper is as follows:1.A task scheduling algorithm for a single MEC server is proposed.This paper combines the characteristics of MEC server and several traditional task scheduling algorithms to model and solve the task assignment problem by quadratic sequence programming and branching method.After the task assignment is clear,the task of the MEC server is deeply analyzed,that is,data volume,practical significance,real-time requirements,etc.On this basis,an efficient single MEC server task schedulingalgorithm is proposed based on Johnson rule,first come first serve and multilevel queue scheduling algorithm.2.A task scheduling algorithm between MEC servers is proposed.When a single MEC server is saturated,which needs to unload the received tasks to the surrounding MEC servers.This paper mainly studies several traditional methods of server load balancing,such as polling method,random method and shortest connection number method,and analyzes their shortcomings.According to the characteristics of MEC server,a task scheduling method based on the shortest response delay is proposed.3.Built an Internet of things platform based on MEC server.Based on the GatewayWorker framework,a Internet of things platform based on MEC server is built by using raspberry pie as the MEC server,camera and other IoT devices as the clients,and verify the effectiveness of the policy on the platform.