ISRI Seminar Series



Interest-Based Self-Organizing Peer-to-Peer Networks:
A Club Economics Approach

Ramayya Krishnan
W. W. Cooper and Ruth F. Cooper Professor of Information Systems

Thursday, 28 October 2004, 12 pm, NSH 1305

ABSTRACT

Improving the information retrieval (IR) performance of peer-to-peer
networks is an important and challenging problem. Recently, the
computer science literature has attempted to address this problem by
improving IR search algorithms. However, in peer-to-peer networks, IR
performance is determined by both technology and user behavior, and
very little attention has been paid in the literature to improving IR
performance through incentives to change user behavior.

We address this gap by combining the club goods economics literature
and the IR literature to propose a next generation file sharing
architecture. Using the popular Gnutella 0.6 architecture as context,
we conceptualize a Gnutella ultrapeer and its local network of leaf
nodes as a "club" (in economic terms). We specify an information
retrieval-based utility model for a peer to determine which clubs to
join, for a club to manage its membership, and for a club to determine
to which other clubs they should connect.

We simulate the performance of our model using a unique real-world
dataset collected from the Gnutella 0.6 network. These simulations
show that our club model accomplishes both perform-ance goals. First,
peers are self-organized into communities of interest - in our club
model peers are 85% more likely to be able to obtain content from
their local club than they are in the current Gnutella 0.6
architecture. Second, peers have increased incentives to share content
- our model shows that peers who share can increase their recall
performance by nearly five times over the performance offered to
free-riders. We also show that the benefits provided by our club model
outweigh the added protocol overhead imposed on the network for the
most valuable peers, that our results are stronger in larger simulated
networks, and that our results are robust to dynamic networks with
typical levels of user entry and exit.

The full paper is available from http://ssrn.com/abstract=585345


Bio:

Ramayya Krishnan is the W. W. Cooper and Ruth F. Cooper Professor of
Information Systems at Carnegie Mellon University. He has a B. Tech in
Mechanical Engineering from the Indian Institute of Technology,
Madras, a M.S. in Industrial Engineering and Operations Research and a
PhD in Management Science and Information Systems from the University
of Texas at Austin. He is an International Research Fellow of the
International Center for Electronic Commerce in Korea and a Visiting
Scientist at the Institute for Information Systems at Humboldt
University (Germany). He is faculty chair of the university's Masters
of Information Systems Management program.

Krishnan's current research interests lie in problems that arise at
the interface of technology, business and policy aspects of
internet-enabled systems. His current research projects investigate
the emergence of virtual communities in peer-to-peer networks, study
intermediation in e-markets and the design of policies that take into
account the competing needs of promoting data access and protecting
privacy. He has published widely on these topics and his work received
the Best Paper awards at the HICSS conference (1997), the Workshop on
Information Technology and Systems (1996, 2000) and the AIS conference
(2001) and was a runner up at ICIS (2002). His work has been funded
over the last decade by the National Science Foundation, The Army
Research Office, and ARPA.