AB-Chord: an efficient approach for resource location in structured P2P networks

Recently, P2P (Peer-to-Peer) technology has witnessed a rapid development. Basically, one of key components in successful P2P applications is how to efficiently look up resources. Considering that structured P2P is a relatively efficient way to locate resources, this paper conducted two improvements to increase the search efficiency in Chord-based algorithms, one of the most popular structured P2P resource lookup protocols. In detail, our contributions are twofold. First, considering the fact that routing information in Chord is not abundant enough for efficient resource search, and looking up resource can only be enforced in clockwise direction, a new algorithm called AB-Chord is proposed to reconstruct the finger tables in Chord, in which counter-clockwise finger table is added to achieve resource queries in both directions, and the density of neighboring fingers is increased. Additionally, AB-Chord implements a new operation to remove the redundant fingers introduced by adding fingers in AB-Chord. Experimental results show that AB-Chord's query efficiency has been improved in terms of the average lookup hops and average lookup delay. And furthermore, considering that the proposed AB-Chord algorithm enlarged the finger table which may cause the forwarding-storm of routing maintenance messages, we further propose AB-Chord+, which appropriately extends the periodic time of updating finger tables and makes the joining and leaving nodes actively send updating messages, to reduce the number of messages forwarded in the network. Simulated results show that AB-Chord+ reduced the network bandwidth consumption.