A popularity based content eviction scheme via betweenness-centrality caching approach for content-centric networking (CCN)

Abstract

In distinction to today’s IP-based, host-bound, Internet architecture, content-centric networking (CCN) emphasizes content by making it instantly addressable and routable. CCN has attracted attention in the research community as a means to cope with the increasing rate of Internet traffic. The host-to-host architecture is demonstrated to be inefficient in content distribution with a lot of bandwidth waste, and it is intricate to set up the network service because of the TCP/IPs location-dependence. CCN is a future Internet architecture which is directed to disentangle the above problems by location-independent content naming and world-wide content caching in a content router. An efficient caching is essential to reduce the delay and to enhance the performance of the network. In addition, a good cache replacement scheme is also necessary to decide which content should reside in the cache and which one should be evicted. The traditional caching replacement strategies schemes such as FIFO, LRU and MRU etc. are not updated as adaption of CCN from host-to-host Internet architecture. Before making replacement, a replacement strategy must performs the calculation of popularity of a content, local popularity of a content (intra-domain network), expected probability respective of demand for a content and instantaneous hit ratio. The traditional and existing popularity based cache replacement strategies do not consider mentioned key points. Therefore in this paper, we present a novel popularity based content eviction scheme for CCN with evaluation of local popularity of a content using the betweenness-centrality concept. The simulation results recommend that our proposed scheme can reliably accomplish the better performance across the other approaches proposed in this field.