Cost effective file replication in P2P file sharing systems

The peer-to-peer file sharing systems is the sharing of files without the interference of the central server. The sharing of files is in need of file replication due to the increase of hot spots. For example, if many clients request the same file to the server at a time, then the server gets overloaded and cannot respond to all the clients. In this situation file replication is needed to get the response with less delay. Current file replication methods replicate files in all nodes or any two end points which in turn results in underutilized replicas. Though file replication takes place in server side, client side and query path solve some problems in current methods but it has high cost of overhead due to more replicas and underutilized replicas. This paper presents an efficient file replication algorithm that achieves high query efficiency and high replica utilization at a significantly low cost. This can be achieved by selecting hot spots, frequent requesters as replica nodes, dynamically adapting to non uniform and time varying file popularity. Simulation results demonstrate the efficiency and effectiveness of this algorithm in comparison with other approaches.