Research On Cache Optimization Algorithm In Fog Computing Network

With the rapid development of mobile Internet,the number of terminal devices connected to the Internet is experiencing explosive growth.Therefore,the problems generated by existing cloud computing are becoming more and more obvious when dealing with such massive requests.The concept of fog computing has been first proposed by Cisco in 2011.The fog node can quickly calculate and store the user's needs,and cooperate with the remote cloud computing to provide more powerful computing ability.Taking into account the characteristics of fog computing that the fog nodes are close to the user,the fog nodes cache the same user's request to reduce the load on the core network.Due to the limited storage capacity and computing power of the fog node,the cache space of fog node needs to be reasonably optimized to meet more user terminal requests.In this thesis,passive and active cache optimization studies are carried out on the fog nodes.The main work of this thesis is as follows:This thesis proposes a link minimum cost cache optimization algorithm based on the Cournot game.Firstly,the cache space of fog node is divided into cooperative and non-cooperative parts.Then,the cache replacement is performed according to the cache value of the content requested by users.Finally,the dynamic Cournot game algorithm is used to bid for the cooperation of each fog node to obtain the maximum benefit,which achieves the minimum link overhead of the entire cache system.The simulation results show that the proposed algorithm is superior to the traditional cache algorithms in terms of cache hit rate and link cost.This thesis proposes a dynamic cooperative cache algorithm based on user mobility.Firstly,the traditional hidden Markov prediction model is used to predict the location information of the user terminal.Then,the autoregressive integral sliding model is used to cache the requested content beforehand on the fog nodes along the user path,in order to optimize the cache utilization of the entire cache system.Finally,the simulation results show that the proposed algorithm has improved cache hit rate and network traffic.