Research On Intelligent Service Scheduling Mechanism Based On Edge Computing In Optical Access Network

With the rapid development of 5G network and Internet of things technology,new applications such as virtual reality,electronic medical,driving assistance and face recognition put forward higher requirements for low delay and high reliability of the network.However,in the traditional network architecture,services can only be transmitted to the core network for centralized processing.Long transmission distance and large bandwidth occupation make cloud computing unable to meet the performance requirements of the new application with delay sensitive characteristics.In this case,edge computing arises at the historic moment.By sinking the computing and network resources to the user's edge side,terminal tasks can be unloaded to the nearest edge computing server for execution,which effectively reduces the end-to-end delay of tasks.Optical access network is a bridge between users and core network.Edge cloud is deployed on the convergence side of optical access network or near the base station,which can support time delay sensitive services to a large extent and provide users with better service guarantee.However,the optical access network supporting edge computing is also facing some problems and challenges.On the one hand,there are more and more access terminal devices,and the high concurrent requests of users will lead to high task load in the network,which will have a certain impact on the processing of time delay sensitive services;on the other hand,due to the cost problem,the resources in a single distributed edge computing server are limited.This makes it very likely to become a new target of DDoS attack,and there is no relevant mechanism to defend.In view of the above problems,this paper studies how to guarantee low delay service delay in high load network and reduce the impact of DDoS attacks on delay sensitive services,and proposes corresponding intelligent service scheduling mechanisms to effectively guarantee the user's experience quality.The specific research of this paper is as follows:In order to meet the needs of various types of services,the ONUs and OLT in the optical access network are deployed with edge computing servers.At the same time,considering that there are many services,the communication between ONUs will adopt centralized and distributed transmission modes,so as to increase the flexibility of service scheduling in the system.Aiming at the problem of high load in optical access network supporting edge computing,an end-to-end delay model of service is established,and an intelligent service scheduling mechanism based on edge cooperation is proposed.In this algorithm,edge nodes can cooperate to process services.According to the analysis of simulation results,this mechanism balances the services load in the network and effectively reduces the blocking rate and QoS degradation rate of delay sensitive services.In the optical access network based on edge computing,an adaptive intelligent traffic scheduling mechanism is proposed for DDoS attack scenarios.The mechanism can be divided into two sub problems:task unloading and resource allocation optimization.The service delay requirements and the resource state of edge computing server in network are considered.Finally,the performance of the proposed algorithm is verified by simulation.In the simulation,two kinds of DDoS attack modes,direct attack and indirect attack,are simulated.Meanwhile,the influence of traffic load,number of attack nodes and attack duration on the index is considered.The results show that the proposed mechanism can effectively reduce the impact of DDoS attacks on delay sensitive services.