An object replacement strategy for global performance in distributed Web caching

The novel feature in the distributed Web caching system we developed is that every client node could act as a cache server and share its cache with neighboring nodes. In this paper we address the problem of object replacement in distributed Web caches. The problem is a version of the knapsack problem which is known to be NP-hard. We propose a heuristic based approach to the problem termed distributed Web caching for global performance (DWCG). Using event driven simulations, we show that in a distributed Web caching environment, the proposed strategy can outperform conventional uncooperative caching strategies such as LFU, LRU and GDSF in terms of LSR, HR and BHR.