Research On SDN-NDN Collaborative Caching Strategy Based On Grey Wolf Optimization Algorithm

With the development of Internet technology,many applications which are based on streaming media technology is becoming mainstream of Internet applications,such as video on demand,live webcast,video chat.Streaming media is more and more being paid attention from researchers,which is to promote the reform and development for the whole internet directly.Named Data Networking(NDN)is a network architecture based on the content-centric.In NDN,each node has some space for cache,the user request could be serviced by the node that what is cache the video.Caching could reduce user access latency significantly,and also it could reduce the traffic flow for cross-network,and reduce server load.Therefore,how to make a caching strategy is one of the key aspects of NDN.Currently,NDN is used ubiquitous cache widely,the cache content could be only have the visibility in the path,which is caused a large amount of redundancy in the network easily and can not make full use of the cache space.To solve this problem,this paper proposes a cache strategy in an autonomous domain and focus on the following three aspects:cache architecture design,cache aware routing and caching strategy.The cache architecture is designed to prepare for making centralized routing decisions and caching strategies;the cache aware routing mainly studies how to design the routing to make full use of the content that has been cached;the main problem is how to develop a reasonable cache policy to optimize the cache performance.In order to solve the above three problems,the following work has been done:(1)In order to collect the information about the network topology and the popularity of the content,we propose a SDN-NDN cooperative caching architecture based on the idea of global control of SDN and then design a routing scheme based on SDN-NDN to improve the utilization efficiency of cache content.Then,we propose a global cooperative caching model to minimize the delay of the user requests for the popular video,which is an integer programming problem,which belongs to the NP-hard problem.(2)In order to solve the NP-hard problem,the grey wolf optimization algorithm is introduced,and two binary format improved methods are proposed.The simulation results show that the algorithm has a good convergence precision and convergence speed.Secondly,we propose a reservation cooperative cache algorithm to solve the global cooperative caching model by using the binary grey wolf optimization algorithm and greedy algorithm.(3)In order to verify the performance of the RCC algorithm,this paper uses the CVX optimizer to solve the cooperative cache model.The experiments show that RCC can obtain the approximate optimal solution with less time and memory.Compared with LCE,ProbCache and OPT-GA caching strategy,the cache strategy of RCC is proved to have higher cache hit ratio and fewer user request video delay.