On sizing CCN content stores by exploiting topological information

In this work, we study the caching performance ofContent Centric Networking (CCN), with special emphasis on thesize of individual CCN router caches. Specifically, we considerseveral graph-related centrality metrics (e.g., betweenness, closeness, stress, graph, eccentricity and degree centralities) to allocatecontent store space heterogeneously across the CCN network, andcontrast the performance to that of an homogeneous allocation.To gather relevant results, we study CCN caching performanceunder large cache sizes (individual content stores of 10 GB),realistic topologies (up to 60 nodes), a YouTube-like Internetcatalog (108 files for 1PB video data). A thorough simulationcampaign allow us to conclude that (i) , the gain brought bycontent store size heterogeneity is very limited, and that (ii) thesimplest metric, namely degree centrality, already proves to bea “sufficiently good” allocation criterion.On the one hand, this implies rather simple rules of thumb forthe content store sizing (e.g., “if you add a line card to a CCNrouter, add some content store space as well”). On the otherhand, we point out that technological constraints, such as line-speed operation requirement, may however limit the applicabilityof degree-based content store allocation