Histogram-Based Global Load Balancing in Structured Peer-to-Peer Systems

Abstract

Peer-to-peer (P2P) systems have rapidly grown in popularity and have become a dominant means for sharing resources in the past few years. Since the nodes are often heterogeneous in these types of systems a load balancing key was a challenging. While several load-balancing schemes have been proposed in the literature, these solutions are typically ad hoc, heuristic based, and localized. In this project, we present a general framework, HiGLOB, for global load balancing in structured P2P systems. Each node in HiGLOB has two key components: 1) a histogram manager maintains a histogram that reflects a global view of the distribution of the load in the system, and 2) a load-balancing manager that redistributes the load whenever the node becomes overloaded or underloaded. A P2P network is partitioned into non overlapping regions corresponding to histogram bucked by exploiting the routing metadata. A mechanism was proposed to keep the cost of constructing and maintaining the histograms low. This scheme can control and bound the amount of load imbalance across the system. Skip Graph, BATON, and Chord are the three existing structured P2P system where we demonstrate the effectiveness of HIGLOB by instantiating. Experimental results clearly indicate that our approach works well.