Computing Offloading And M-Trail Failure Localization In Edge Clouds

With the increasing popularity of smart mobile terminals,the demand of mobile users for intensive computing application services has shown an exponential growth trend.Traditional wireless cellular networks are unable to meet this exponential growth both in high data rates and high computing power.The combination of communication and computing power has become more important in wireless networks,giving rise to the concept of mobile edge computing.The main idea is to offload the task from the mobile devices to the cloud.And then the task access to the edge of the network through the wireless network.Users can obtain efficient computing resources and get closer to the computing resources.This technology can solve the problem of limited computing and storage capacity of mobile terminals,reduce the communication delay of mobile terminals and save energy consumption,thus improving user experience.Mobile edge computing promotes the development of mobile interconnection communication,but it still faces many challenges in the aspects of unstable wireless network connection,network heavy load and sensitive business delay.Aiming at the above problems,this paper mainly studies on computing offloading in mobile edge clouds.We proposes a model for distribution placement of multi-task project,through jointing by computing infrastructure(computing resources,storage resources),the time-varying wireless channel between the mobile user and base station.The taskes are placed in the base station of the virtual machine.The target is to minimize the service cost from the veiw of the cloud service provider,ensuring the quality of service to a certain extent.We use Integer Linear Program(ILP)to solve this problem,and propose corresponding heuristic algorithm MAGA(multi-population adaptive genetic algorithm)based on genetic algorithm.The simulation results show that ILP is correct and the heuristic algorithm is effective.In order to improve the stability and fault tolerance of data center network in edge cloud,this paper also introduces the idea of monitoring trail.Under the Clos architecture of the data center in edge cloud,a method of fault monitoring and rapid localization is proposed to realize fault monitoring and localization of links and switching equipment.The goal is to minimize the monitoring system cost.In this paper,the problem is modeled and optimized by ILP.Simulation results verify the validity of the proposed ILP model.