Gossip-based search selection in hybrid peer-to-peer networks: Research Articles

Authors:
M. Zaharia;S. Keshav
Affiliations:
School of Computer Science, University of Waterloo, Waterloo, ON, Canada;School of Computer Science, University of Waterloo, Waterloo, ON, Canada
Venue:
Concurrency and Computation: Practice & Experience - Recent Advances in Peer-to-Peer Systems and Security (P2P 2006)
Year:
2008

Citing 0
Cited 6

Evaluation of a selective distributed discovery strategy in a fully decentralized biologically inspired environment

Information Sciences: an International Journal
Demand-driven publish/subscribe in mobile environments

Wireless Networks
Review: A survey on content-centric technologies for the current Internet: CDN and P2P solutions

Computer Communications
Cooperative stalking of transient nomadic resources on overlay networks

Future Generation Computer Systems
The state of peer-to-peer network simulators

ACM Computing Surveys (CSUR)
Gossip-Based Self-Organising Agent Societies and the Impact of False Gossip

Minds and Machines

Quantified Score

Hi-index	0.00

Visualization

Abstract

We present GAB, a search algorithm for hybrid peer-to-peernetworks, that is, networks that search using both flooding and adistributed hash table (DHT). GAB uses a gossip-style algorithm tocollect global statistics about document popularity to allow eachpeer to make intelligent decisions about which search style to usefor a given query. Moreover, GAB automatically adapts to changes inthe operating environment. Synthetic and trace-driven simulationsshow that compared to a simple hybrid approach that always floodsfirst, trying a DHT if too few results are found, GAB reduces theresponse time by 2550% and the average query bandwidth cost by 45%,with no loss in recall. GAB scales well, with only a 7% degradationin performance despite a tripling in system size. Copyright ©2007 John Wiley & Sons, Ltd.