Performance Analysis And Simulation On Cache Mechanism Of The Future Network

With the rapid development of Internet technology, as well as the increasing number of Internet users,the traditional end-to-end communication mode is becoming more and more difficult to meet the needs of users, particularly reliable transmission of large quantities of content and high quality services. In order to solve the fundamental problem, recent years, researchers have proposed a series of future network architectures. One common and important feature of these architectures is to use caching mechanism, Caching mechanism can not only improve the hit rate and reduce the delay of user request, but also reduce network traffic, the study of cache mechanism is becoming increasingly important.In this thesis, the two implementations of future network-information-centric network and delay-tolerant network caching mechanisms are analyzed. In Information-Centric Network(ICN), this thesis focuses on a chunk caching location and search scheme(CLS), by adding a cache trail which is used to record the data copy of the content in the router, to ensure that there is only one copy of the content on transmission path, while optimizing off-path searching. Comparing the number of hops between current node and the content server to a threshold set before, which used to decide weather forward the request to upstream or downstream. Finally, the simulation experiments show that, CLS is better than conventional caching strategies on caching hit ratio, caching hit distance, average download latency and bandwidth consumption.In Delay-Tolerant Network(DTN), due to intermittent connection, DTN use“storage-carry-forward” to transfer data, routing algorithm is the key to DTN research,while caching can greatly affect the routing algorithm. This thesis analyzes a caching management based on message transmission status, the scheme can suppress the further spread of the message which has more copies and spreads faster, giving more opportunity to the message which has less copies and spreads slower in the network.The simulation results show that the caching management is better than other commonly used caching strategies on delivery ratio, overhead ratio and average delay.