Socially-aware NodeRank-based caching strategy for Content-Centric Networking

Content-Centric Network (CCN) emerges as a promising network architecture. In CCN, each node is cache-equipped, so one of the most important issues is how to cache content chunks in different networks to optimize the network performance. Meanwhile, the Internet is gradually becoming a social-oriented network. Therefore, it is essential to include social characteristics into the network traffic model. Social links have a direct influence on the network delivery efficiency and we can utilize social information to lighten network load. This paper proposes a novel caching strategy based on nodes' importance to social communities in CCN. We utilize social information and spectrum of graph to divide the network into several social Autonomous Sections (sociAS). In each sociAS, we choose the influential node to cache popular contents according to node's rank. By this means, the node with the highest rank can pro-actively cache contents and distribute contents to others. Different from other caching schemes, we combine physical attribute of network with users' social behaviors to realize caching decisions. Extensive simulation results demonstrate that our scheme achieves better caching performance, including: cache hit ratio, cache hit distance, delay of download and server load saving.