An efficient ranking-matched caching strategy for information centric networking

Li Ding,Jinlin Wang,Yiqiang Sheng,Lingfang Wang
DOI: https://doi.org/10.24507/ijicic.14.02.519
2017-01-01
Abstract:The host-based communication model of current Internet infrastructure is proved to be inefficient in content distribution because of a lot of bandwidth waste. In this context, Information Centric Networking (ICN) emerges as a promising future architecture that aims to distribute content efficiently over the network. The key to efficient content distribution in ICN is in-network caching, which is an effective way of eliminating redundant network traffic. Cache decision at each router is a very important factor which governs caching performance. In this paper, we propose a content Popularity Ranking and node Importance Ranking Matched (PRIRM) caching strategy for ICN, which is targeted at reducing cache redundancy and improving content diversity. Content popularity represents the user preference while node importance determines the important degree in the network. By disseminating content on the delivery path based on the popularity ranking of the content and the importance ranking of the node, content with different rankings can be distributed in the network hierarchically. The sensitivity of PRIRM strategy against different parameters, such as content number, popularity distribution and network topologies, is thoroughly studied. The simulation results show that PRIRM strategy can result in a significant decrease of average hop count, compared with three benchmark caching strategies, i.e., leave copy everything strategy, a centrality-based strategy and a path-capacity-based strategy.
What problem does this paper attempt to address?