Research On Resource Scheduling And Offloading Strategies In Edge Computing

At present,mobile data traffic is in a very fast growth stage due to a large number of new services.In the new era of data mining,large amounts of data to be processed will be collected by sensors or IoT devices.Some applications need to process task data within a very short delay,but ordinary sensors are difficult to run complex programs,and the delay of data re-uploading to the cloud is large,sometimes far greater than the low delay constraint of some applications,thus affecting the quality of service of tasks.In order to overcome this problem,the edge of mobile computing(MEC)on the edge of the network is introduced into the calculation function,it aims to ensure effective network operation and service delivery on the basis of effective reduce latency,and the content and services to users,make applications,services,and content are deployed closer to the user's local,so as to provide users with low latency and high bandwidth and real-time service environment.Mobile Edge Computing(MEC)greatly improves the quality of the computing experience by transferring computation-intensive workloads to MEC servers,but computational offloading strategies face new challenges due to the competing execution of resources and the optimal selection of servers during offloading.The main research contents of this paper are described as follows:(1)Based on the task unloading problem in multi-server scenario,a server region division algorithm was proposed to reduce transmission energy consumption,and the algorithm was applied to edge computing task unloading to achieve a trade-off between energy consumption and overall task execution time and minimize the total task execution cost.The problem was modeled as the joint optimization problem of task allocation and offloading,and the energy consumption problem of offloading based on server region division was considered.The offloading decision problem among mobile device users was transformed into the multi-user game problem.By realizing Nash equilibrium,the optimal solution of multi-user task offloading problem was obtained.Simulation results show that the algorithm can effectively reduce the energy consumption of task execution and improve system performance.(2)We study a green MEC system with EH equipment and propose an effective computing unloading strategy.By selecting the execution mode between the local execution,unloading execution and task discard of each mobile device,the algorithm can progressively obtain the optimal result of the whole system.The simulation results validate the theoretical analysis and the effectiveness of the algorithm.On the premise of ensuring QoE,this algorithm can effectively improve the load rate of computing tasks.