Optimal P2P Cache Sizing: A Monetary Cost Perspective on Capacity Design of Caches to Reduce P2P Traffic

Abstract

Peer-to-Peer (P2P) systems are generating a large portion of the total Internet traffic and imposing a heavy burden on Internet Services Providers (ISPs). Proxy caching for P2P traffic is an effective means of reducing network usage, thereby reducing operation costs for ISPs. Proxy cache storage design has a significant impact on ISPs. While there are several works on how to optimally design cache locations and capacity allocation to each location given a total capacity, few works tell ISPs what is the optimal total P2P cache storage capacity. In this paper, we propose an analysis method to the problem of optimally determining P2P cache size. An analysis methodology is proposed to determine the optimal cache size by considering the monetary costs of cache storage and bandwidth. Guided by our model, a close-form expression is developed to guide an ISP in the cache capacity design. Numerical evaluation results show that ISPs can achieve significant cost saving by deploying P2P cache and by allocating the cache capacity optimally.