Load Balancing in Peer Networks

Abstract

Peer to peer systems have emerged as an appealing solution for sharing and locating resources over the Internet. Several P2P systems have been successfully deployed for a wide range of applications, such as Gnutella, Bit Torrent and Skype (Internet telephony system). A Peer makes a portion of its resources such as processing power, disc storage or network bandwidth directly available to other network (peers) without the need for central coordination by servers or stable hosts. Load balancing aims to optimize resource use, maximize throughput, minimize response time, and avoid overload of any single resource. One of the most commonly used applications of load balancing is to provide a single Internet service from multiple servers, sometimes known as a server farm. We are evaluating the performance of load balancing strategies in structured overlay networks. This study discusses the pros and cons of the load balancing strategies in structured overlay network. The goal of our project is to select high performing load balancing strategy to evenly divide the load among the processing elements. It is critical to design a P2P system that is scalable and efficient. To build an efficient P2P system, researchers have turned to structured architectures which offer a bound on search performance as well as completeness of answers. However, one key challenge that has not been adequately addressed in the literature is that of load balancing. In a large-scale P2P system, nodes often have different resource capabilities. Hence, it is desirable that each node has a load proportional to its resource capability. We implement a framework, called Histogram-based Global LOad Balancing (HiGLOB) to facilitate global load balancing in structured P2P systems. In this paper, 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 under-loaded.