Load Balancing Model Based On Network Positioning In Structured P2P Systems

With the great development of PC capabilities and the fast growth of Internet users, there exists a vast quantity of computing and storage resources on the edge of the Internet. And network bandwidth is greatly improved, which makes it possible to exploit and use these potential resources. How to effectively use these rich computing resources has become a hot issue. P2P technology is proposed to meet the requirments.P2P is a distributed system, in which nodes share their part of resources. These resources can be shared directly without any intermediate entities, which makes communication on the Internet become easier and more direct. And it changes the state that large scale servers were the center on Internet。In peer-to-peer systems, the key issue to distribute real-time data is load balancing because video and audio traffic is increasing on the Internet, especially on peer-to-peer networks. When the traffic is crowded in a smaller LAN, it is likely to cause congested, or even collapse. In addition, the heterogeneity of node capacity and semantic relativity of keys can cause load imbalance among nodes.In this paper, we research the problem of load balancing in structured P2P systems. Based on Chord, a load balancing scheme is proposed according to network positioning technology. On the one hand, we improve the network positioning technology and introduced it to the DHT (Distributed Hash Table) network. Then, load transfer can achieve among nodes physically close and the consumption of bandwidth and delay is minimized. On the other hand, we use an area transfer method as the load transfer strategy. When a light node arises, the node, which is work as a center, will construct a star-like structure area with other nodes physically close to it. Load transfer will achieve in a neighboring region. Regional load transfer method can accelerate the spread speed and reduce delay.This paper describes the improvement and application about these two aspects in detail, and introduces the simulation environment, experiment process and experiment results of the load balancing scheme. .