An efficient and social-aware distributed in-network caching scheme in named data networks using matching theory

Abstract

One of the main characteristics of Named Data Networks (NDN) which makes it a promising architecture for the future of the Internet is its universal in-network caching. Due to the limited caching capacity of the network and the infinite amount of data objects, the cache management scheme becomes a challenging task in NDN. While existing cache management schemes utilize network’s caching capacity inefficiently by redundantly keeping popular objects across the network and concentrate on improving bandwidth utilization, we focus on a novel criterion to elevate the cache utilization efficiency. In this paper, a distributed solution that associates the objects to the cache-enabled nodes and maximizes the total network’s social welfare is proposed which models the problem of network’s social welfare maximization as a many-to-one matching game. Further, it is proved that the proposed solution has a number of desired properties. The resulting matching is stable and the outcome of the proposed algorithm is in Competitive Equilibrium, the social-welfare of the proposed algorithm converges to the social-welfare of the centralized optimal approach as the price step number is sufficiently small, the proposed algorithm converges after a finite number of iterations with relatively low overhead and the cost of retrieving the required objects is the minimum for the Internet Service Providers (ISP). Finally, simulation results show the convergence of our proposed scheme to the centralized optimal solution and its outperformance against the default applied caching scheme in NDN in terms of both network’s social welfare and hit ratio.