A value-based cache replacement approach for Information-Centric Networks

Information-Centric Networks (ICNs) represent a content-based model which addresses user´s requests regardless of the content´s location or the nature of its original publisher. The performance of an ICN is highly dependent on replicating the content across the caches of a multitude of nodes in the network. Given the high data turnover rates of contemporary applications and the finite nature of caching space, efficient caching algorithms play a crucial role in determining which data item can be safely dropped in order to accommodate for more important items. In this paper, we present a value-based cache replacement approach that executes a Least Valuable First (LVF) policy. Our approach employs a utility function that uses delay, popularity and age parameters to determine which item to drop from the cache. We present simulation results comparing our approach to other dominant cache replacement policies under varying conditions such as data popularity, in-network cache ratio and connectivity degree. Results show that our approach outperforms in terms of time-to-hit, hit rate, in-network delay and data publisher load.