Research On Task Offloading Scheduling Method Based On Edge-Cloud Cooperation

With the rapid development of 5G technology,more and more wearable devices and other mobile Io T devices have become very popular in modern society,and the data generated by these terminal devices is increasing exponentially.Although cloud computing has powerful processing power,the network delay is relatively large due to the distance from the terminal device.Mobile edge computing(MEC),as a new computing paradigm,is close to smart terminal devices,but cannot meet the needs of users due to resource constraints.In view of this situation,this paper considers the complementary advantages of the two through edge-cloud collaborative computing to improve the quality of service for users.The main research contents of this paper are as follows:First,cloud computing is introduced into the "device-edge" network framework of MEC,and a "device-edge-cloud" three-layer collaborative computing network architecture with multiple users,a single MEC,and a cloud computing center is constructed.Under this framework,the offloading and scheduling problem of tasks are studied.In order to improve the efficiency of "device-edge-cloud" collaborative computing and serve as many users as possible,a task offload scheduling algorithm(RROS)based on response ratio is given.The corresponding offloading decision is determined according to the task’s consumption function and deadline in different situations,then the priority of the task to be processed is dynamically adjusted based on the response ratio of the task so that more users can participate in the offloading of the task.Secondly,this paper also considers the situation of multiple MECs serving users,that is,to build a "device-edge-cloud" three-layer collaborative computing network architecture with multiple users,multiple MECs,and a cloud computing center.Under this framework,it is considered that some users will have a certain degree of preference for MEC services based on their previous experience with different MECs in terms of service quality and charging standards.At the same time,in order to improve the efficiency of "device-edge-cloud" collaborative computing and serve as many users as possible,a task offloading and scheduling algorithm(UPRROS)based on user preference is given.The algorithm increases the user’s selectivity when offloading tasks by introducing a user preference factor and dynamically adjusts the tasks on different MECs based on the task response ratio,further improving the task offloading rate and serving more users.Finally,the performance of the RROS and UPRROS algorithms is evaluated through simulation experiments.The experimental results show that the algorithm can provide users with effective offloading strategies,allowing more users to participate in offloading and saving costs,thereby improving the quality of service to users.