Research On Offloading And Migr Ation Method And Application Of Fdge Cloud Task

With the rapid development of Internet of vehicles and Internet of things technology,the number of various terminal devices,including mobile phone terminals,automobile terminals and small computing terminals,has increased explosively;In particular,in recent years,the rapid development of technologies such as autonomous driving,virtual reality(VR),augmented reality(AR)and machine learning,as well as the continuous introduction of concepts such as metaverse and blockchain,have led to a large increase in the demand for computing services on the terminal side,boosting the rapid development of edge cloud technology.In order to adapt to the changes of data services and network quality(QoS),the computing resources and computing power of the whole edge cloud can be brought into full play through the method of edge cloud task offloading and migration,in order to improve the overall effect of edge cloud services.Therefore,the offloading and migration strategies,methods and applications of edge cloud tasks need to be deeply studied.Firstly,this paper maps edge cloud tasks to container cloud environment to transform the task migration problem into container migration problem,and proposes a migration algorithm based on ACO algorithm and tabu search algorithm fusion(ACO-TS).Then,the Cloudsim simulation program is used to simulate and test the algorithm.Compared with the classical max-min,ACO algorithms and so on,the results verify that the proposed algorithm is more efficient and faster,and in consideration of the performance of edge clustering load balancing.Finally,based on Docker container cloud platform and Kubernetes container scheduling technology,a set of edge cloud offloading and migration system is independently developed and implemented according to ACO-TS migration algorithm.The system senses the status of edge cloud nodes and automatically triggers the offloading and migration strategy when a node edge cloud computing becomes overloaded.At the same time,the monitoring function of user edge clustering environment is also added,and the system performance test results reach the actual target of the system.