Accession Number:

ADA455758

Title:

AGNO: An Adaptive Group Communication Scheme for Unstructured P2P Networks

Descriptive Note:

Technical rept.

Corporate Author:

MARYLAND UNIV COLLEGE PARK DEPT OF COMPUTER SCIENCE

Report Date:

2004-01-01

Pagination or Media Count:

17.0

Abstract:

We present the Adaptive Group Notification AGNO scheme for efficiently contacting large peer populations in unstructured Peer-to-Peer networks. AGNO defines a novel implicit approach towards group membership by monitoring demand for content as this is expressed through lookup operations. Utilizing search indices, together with a small number of soft-state shortcuts, AGNO achieves effective and bandwidth-efficient content dissemination, without the cost and restrictions of a membership protocol or a DHT. We present several simulation results over different topologies using both synthetic and real workloads, to evaluate its performance. Our method achieves high-success content transmission at a cost at least two times smaller than proposed techniques for unstructured networks.

Subject Categories:

  • Computer Systems

Distribution Statement:

APPROVED FOR PUBLIC RELEASE