(英) |
In this paper, we consider large-scale CCN (Content-Centric Networking) networks and derive the cache hit probability at each router, the average content delivery delay for each entity, and the average content delivery delay for the entire network over a content distribution tree which comprised of a single repository (i.e., storage servers) and multiple entities (i.e., content consumers). In the literature, several performance analyses of CCN in a general network topology have been performed. Those performance analyses are effective for analyzing characteristics of small-scale CCN networks. However, it is not easy to utilize those existing performance analyses for analyzing large-scale CCN networks since the amount of computation required for those analysis increases according to both of the network size and the number of contents. Therefore, in this paper, we approximately derive the cache hit probability in each router and the average content delivery delay for each entity in large-scale CCN networks. Through several numerical examples, we investigate the effect of the structure and the size of a content distribution tree and the content store size at a router on the average content delivery delay for the entire network. Our findings include that when the content store size at every router is sufficiently large, even if a CCN network becomes large, growth of the average content delivery delay for the entire network can be suppressed. |