Energy Consumption And Utility Optimization Based On Server Collaboration In Mobile Edge Computing

With the rapid growth of the number of smart devices and the increase of delaysensitive applications,cloud computing can no longer meet the needs of such applications.At the same time,due to the limited computing resources and battery capacity on mobile terminals,the European Telecommunications Standards Institute(European Telecommunications Standards Institute,ETSI)proposed the concept of Mobile Edge Computing(MEC)in 2014.Users can use edge servers located near the terminal for task processing,which is conducive to reducing task delay and terminal energy consumption.Although a large number of edge servers deployed at the edge of the network bring convenience to users,the uncertainty of the number of mobile terminals can easily lead to imbalanced load on the edge servers and generate more energy consumption,making resource allocation in mobile edge computing face complex energy consumption and The problem of utility optimization,therefore,how to allocate computing resources at the edge of the network,optimize energy consumption and utility of edge servers,and ensure user service quality are important challenges facing mobile edge computing systems.At present,many scholars have conducted research on this,but the existing research has the following deficiencies:(1)In the study of resource allocation and energy consumption optimization based on a single edge server,it is not considered that the overlapping area of the edge server and other servers will cause the mobile terminal to uninstall The choice is more complicated.At the same time,with the increase of the number of edge servers,more tasks will be offloaded to the edge nodes,which will easily aggravate the degree of network congestion;(2)In the research of resource allocation and energy consumption optimization based on multiple edge servers,there is no Consider the energy waste caused by unbalanced load on the edge server.In view of the above deficiencies,this article conducts research from the following two aspects:(1)An edge server collaboration mechanism is proposed.Among edge servers close to the user,only one edge server acts as a leader and processes tasks for all terminals in the range.However,when the edge server leader has computing resources When insufficient,consider offloading tasks to the remaining edge servers that are collaborator servers.This method integrates network edge resources.Compared with a single edge server,multiple edge server collaboration mechanisms can provide more computing resources and reduce the number of tasks offloaded to the cloud center.Simulation experiments show that this method can keep the task buffer queue on the leader server stable for a long time.(2)An edge server sleep mechanism is proposed.The edge server acting as a collaborator in the system has two states of sleep and active.When the sleep state consumes less energy,the edge server will receive tasks and process the active state.Control the status of collaborators according to the status of the leader.In this paper,the dormancy mechanism is combined with the edge server collaboration mechanism,and the energy consumption utility of multiple edge servers is optimized.Simulation experiments show that this mechanism can ensure the long-term stability of the task buffer queue on the leader's edge server while optimizing energy consumption.