Dynamic Load Balancing with Partial Knowledge of System in Peer to Peer Networks

Load balancing is a critical issue for the efficient operation of peer-to- peer networks. The goal of P2P systems is to harness all available resources (storage, bandwidth, and CPU) in the P2P network so that users can access all available objects efficiently.P2P aims to directly balance the distribution of itemsamong the nodes. With the notion of virtual servers, peers participating in a heterogeneous, structured peer-to-peer (P2P) network may host different numbers of virtual servers, and by migrating virtual servers, peers can balance their loads proportional to their capacities. Peers participating in a Distributed Hash Table (DHT) are often heterogeneous. Potential P2P substrates are based on Distributed Hash Tables. The existing and decentralized load balance algorithms designed for the heterogeneous, structured P2P networks either explicitly construct auxiliary networks to manipulate global information or implicitly demand the P2P substrates organized in a hierarchical fashion. Without relying on any auxiliary networks and independent of the geometry of the P2P substrates, we present, in this paper, a novel efficient, proximity-aware load balancing algorithm by using the concept of virtual servers, that is unique in that each participating peer is based on the partial knowledge of the system to estimate the probability distributions of the capacities of peers and the loads of virtual servers, resulting in imperfect knowledge of the system state. With the imperfect system state, peers can compute their expected loads and reallocate their loads in parallel.