Improving Lookup Performance and Consistency in DHT

Abstract

During recent years, Distributed Hash Table (DHTs) have been extensively studied through simulation and analysis. As a fundamental problem in DHT based P2P system, lookup consistency and load balancing is important to avoid performance degradation and guarantee system fairness. In this paper, three important aspects: lookup consistency, group communication and lookup replication are involved to improve DHT lookup performance and load balancing. Lookup consistency maintains a ring structure and guarantees consistent lookup results in the presence of node join and leave, regardless of where the loop is initiated. Broadcast algorithm for group communication avoids redundant messages. Lookup replication places replicas in a DHT called Symmetric replication that enables parallel lookup results. Parallel lookups are known to reduce latencies. A new system will be implemented by integrating the three algorithms in a middle ware.