Hierarchical Bidirectional Chord

Abstract

In peer-to-peer (P2P) networks, how to effective locate data on distributed hash tables is challenging and has attracted much research attention in recent years. Two measurements are usually considered when discussing the effective location of data items: the path length (the hop count for resolving a lookup request on the overlay network) and the latency (the actual time period between the issuing and completion of a lookup). In this paper, we propose a regional lookup algorithm based on super-node (Hierarchical DHT Algorithm) that routes data keys queries by region information and super-node, and keeps the finger tables on two directions to reduce the redundancy of finger tables in original Chord. The lookup process can be easily implemented by keeping only O (log2M) routing information with M regions in the Chord ring at each node. Results from theoretical analysis and simulations show that the improved lookup algorithm achieves higher lookup efficiency and scales well.