Decision-making for in-network caching of Peer-to-Peer content chunks - An analytical modelling study (original) (raw)
2014, 2014 International Conference and Workshop on the Network of the Future (NOF)
ABSTRACT In-network content caching has recently emerged in the context of Information-Centric Networking (ICN), which allows content objects to be cached at the content router side. In this paper, we specifically focus on in-network caching of Peer-to-Peer (P2P) content objects for improving both service and operation efficiencies. We propose an intelligent in-network caching scheme of P2P content chunks, aiming to reduce P2Pbased content traffic load and also to achieve improved content distribution performance. Towards this end, the proposed holistic decision-making logic takes into account context information on the P2P characteristics such as chunk availability. In addition, we also analyse the benefit of coordination between neighbouring content routers when making caching decisions in order to avoid duplicated P2P chunk caching nearby. An analytical modelling framework is developed to quantitatively evaluate the efficiency of the proposed in-network caching scheme.
Related papers
Caching and relaying strategies for peer-to-peer content delivery
2007
— Peer-to-peer content distribution has become a major source of bandwidth costs for Internet service providers (ISPs). One way for ISPs to decrease these costs is to deploy caches for p2p traffic. To make efficient use of the caches, in this paper we propose a cooperative caching and relaying scheme that is compatible with the existing business relations between ISPs. We formulate the problem of cooperative caches as a resource allocation problem, and show that it is related to the problem of r-configuration studied in graph theory. We propose a distributed algorithm to solve the resource allocation problem, and show that cooperation leads to significant gains compared to non-cooperative caching. I.
Loading Preview
Sorry, preview is currently unavailable. You can download the paper by clicking the button above.