Optimization of Content-Centric Networks

Today’s Internet has become widely a distribution network where its end users are not concerned with where the data is located, but mainly interested in obtaining the data they want. To evolve the Internet toward its current primary use, Information-Centric Networking (ICN) architectures were developed with in-network caching as one of their key features. However, they have yet to settle on specific traffic engineering and caching strategies. In the first chapter of this thesis, we develop MIRCC, a rate-based, multipath-aware congestion control approach for ICN. We first present MIRCC’s algorithm for single-path flows and develop a non-recursive rate-calculation algorithm which achieves max-min fairness, high link utilization and short flow completion time. We then focus on multi-path flows and design a novel hybrid scheme with dual-class rate management, in which each flow has two rate levels: the primary rate is ensured a level of max-min fairness between all flows and the secondary rate is managed to consume remaining bandwidth resulting in full link utilization. In chapter two, we model and formulate the traffic engineering (forwarding strategy) and in-network caching jointly under ICN architectures into an optimization problem, with the objective of minimizing the network delay. We then solve the problem and develop fully distributed and modular algorithms for both hop-by-hop forwarding and caching strategies. Through extensive simulations, we demonstrate that our forwarding and caching strate-