CPePC: Cooperative and Predictive Popularity based Caching for Named Data Networks

Abstract

Caching content is an inherent feature of Named Data Networks. Limited cache capacity of routers warrants that the choice of content being cached is judiciously done. Existing techniques resort to caching popular content to maximize utilization. However, these methods experience significant overhead for coordinating and estimating the popularity of content. To address this issue, in this paper, we present CPePC, which is a cooperative caching technique designed to improve performance. It accomplishes this through a combination of two factors. First, CPePC enhances efficiency by minimizing the overhead of popularity estimation. Second, it forecasts a parameter that governs caching decisions. Efficiency in popularity estimation is achieved by dividing the network into several non-overlapping communities using a community estimation algorithm and selecting a leader node to coordinate this on behalf of all the nodes in the community. CPePC bases its caching decisions by predicting a parameter whose value is estimated using current cache occupancy and the popularity of the content into account. We present algorithms for community detection, leader selection, content popularity estimation, and caching decisions made by the CPePC method. We evaluate and compare it with six other state-of-the-art caching techniques, with simulations performed using a discrete event simulator to show that it outperforms others.