Content storage effect on the named data network traffic load

Abstract

Named Data Network (NDN) is a network architecture with a content-centric paradigm, in which the consumer has a certain interest in a content. Requests for data are performed based on the content, not to specific server or producer so that the response to the requested content can be performed by the NDN router, not only by the server or producer. NDN causes data traffic more efficient because the request will be served by the node that closest to the consumer who has the content in its Content Storage. In this research, the performance of NDN is tested based on changes in the size of Content Storage as one of the structures in NDN, and also the changes of frequency of interest. The contribution of this research is to study the effect of the content storage size and the number of frequency of interest on NDN performance, from the network point of view and user point of view. This research use Abilene network topology. The performance will be analyzed from average delay and cache hit ratio as the user centric metrics, and also the data circulation, number of interest message retransmission and packet drop as a network centric metrics. Simulation results show that the size of Content Storage greatly affects the efficiency of data transmission in the NDN network. The smaller the size of Content Storage relatively compared to the number of consumer demand, the greater the network load. Data circulation becomes larger, delay and packet drop increases, the number of packets retransmission increases and the cache hit ratio deteriorates. The provision of Content Storage of the optimum size makes the data transmission on the NDN network efficient. The higher number of frequency of interest, the worse the performance of the system.