Research On Task Offloading Strategy Based On Edge Computing For Industrial Internet

With the proliferation in the number of devices involved in industrial internet,it is becoming more and more difficult to meet simultaneously the requirements of industrial applications in terms of latency and economy only using traditional cloud computing paradigm.Edge computing,which is the latest computing paradigm,can provide faster service response and reduce network overhead.However,compared to cloud computing paradigm,when the number and complexity of tasks increase,it may face the problem of unable to the requirements due to resource constrained.In order to ensure the low latency and energy consumption of industrial devices,this paper studies the current status of task offloading in cloud and edge computing scenarios,and analyzes several typical task offloading strategies.On this basis,a cloud and edge computing integrated computing architecture for industrial internet is proposed,which allows industrial applications to offload tasks to edge or cloud servers to ensure quality of service(QoS)of industrial applications.Based on the integrated computing architecture,a formal description of multi-objective optimization problem is given for the offloading problem of the application with delay and energy sensitivity in industrial internet scenarios are proposed.Aiming at the proposed multi-objective optimization problem,this paper proposes two task offloading algorithms ASO and Pro-ITGO.The ASO algorithm is a polynomial time algorithm with three sub-algorithms,including sub-deadline allocation,topology sorting,and task offloading algorithms.The Pro-ITGO algorithm is a group intelligent heuristic algorithm,which improves the four cell update stretagies and fitness function of the original ITGO algorithm(Invasive Tumor Growth Optimization)to adapt the task offloading scenario.In this paper,the task offloading experiments are performed on the two algorithms based on the WorkflowSim simulation environment.Through the evaluation and analysis of the experimental results,relative to the basic algorithms and other heuristic algorithms,the two algorithms have obvious advantages in task completion time,the energy consumption of industrial devices and cloud computing costs.