Research On The Data Caching Optimization In Mobile Internet

As the boost development of mobile devices and increase of global content,content requirement and mobility support have gradually became urgent challenges.However,it is difficult for the traditional internet architecture to handle content and mobility.For one thing,the switch device is the bottleneck of the bandwidth,and the competition of the bandwidth causes network congestion at backbone network.Furthermore,the connection oriented communication mode results in much in-network redundancy,and excessive host centricity becomes the key shortcoming of the network.For the other thing,the dual identity of IP address is inefficient to handle the challenge of mobility,let alone to solve problems like multi-homing.As a result,there has been renewed interest in mobile internet which aims to guarantee QoS of end-to-end services for any user at any time in any place.Various content caching techniques have been adopted in the mobile internet.First,for the mobility,duly caching can not only reduce the loss of throughput caused by link fluctuation,but also decrease the handover delay of mobile nodes,which helps to realize seamless handover.Second,for the content,massive content utilizes in-network caches to decrease point-to-point content transmission in the network,which effectively solves the problem of the redundancy.Hence,this paper studies the optimization of the content caching techniques in mobile internet.This paper focuses on new routing protocol,mobile management and data center of the mobile internet.The main contributions in the paper include:(1)Proposing the strategy of the efficient real-time aware routing in GSTAR.To solve the problems of the sub-optimal path quality and blind caching of real-time packets,this paper proposed an efficient real-time aware routing strategy,which consists of link pre-selection,optimize of forwarding and real-time aware caching.The experiment results illustrates that GSTAR can slightly improve the throughput compared with traditional link state protocol.Compared with GSTAR,the proposed strategy can not only accelerate the forwarding,but also concerns the real-time quality of packets.(2)Proposing enhanced FPMIPv6-based caching strategy.Since FPMIPv6 has the problem of the reduplication of data path and overload of routing interface during handover,this paper designs the new messages for both fast binding and address request,and helps LMA to gain information about MAG to cache data packets at LMA efficiently.Then quantitative analysis of the delay for control packets and data packets during handover is conducted,the results show that enhanced FPMIPv6 can not only solve the problem of overload,but also decrease the handover delay of mobile nodes in the heterogeneous networks.(3)Proposing tendency based caching in data center.Existing strategies lack of the global view,which results in inefficient caches.In this paper,we aim to analyze the real-time distribution of contents among caches from multiple perspectives.This paper proposes TCBRP,a scheme that analyzes caching tendency of various contents in reverse path,based on centrality of nodes,popularity of contents and replacement rate of nodes,to cache in-network content.This scheme also has decent scalability and can be expended conveniently.The experimental results reflect that TCBRP report savings in average hops and balance cache hit rate,compared with BetwRep and LCE.