Bandwidth and storage sharing performance in information centric networking

Internet usage has dramatically evolved towards content dissemination and retrieval, whilst the underlying infrastructure remains tied up to hosts interconnection. Information centric networking (ICN) proposals have recently emerged to rethink Internet foundations and design a natively content-centric network environment. Important features of such networks are the availability of built-in network storage and of receiver-driven chunk-level transport, whose interaction significantly impacts overall system and user performance. In the paper, we provide an analytical characterization of statistical bandwidth and storage sharing, under fairly general assumption on total demand, topology, content popularity and limited network resources. A closed-form expression for average content delivery time is derived and its accuracy confirmed by event-driven simulations. Finally, we present some applications of our model, leveraging on explicit formulae for the optimal dimensioning and localization of storage resources.