Internet Iso-bar: A Scalable Overlay Distance Monitoring System

Yan Chen and Lili Qiu1
(Professor Randy H. Katz)
California Micro, (DARPA) N66061-99-2-8913, Ericsson, Nokia, and Siemens

Estimating end-to-end Internet distance can benefit many applications and services, such as efficient overlay construction, overlay routing and location, and peer-to-peer systems. While there have been many proposals on Internet distance estimation, how to design and build a scalable, accurate, and dynamic monitoring system is still an open issue. To this end, we propose an overlay distance monitoring system, Internet Iso-bar, which yields good estimation accuracy with much better scalability and far less overhead for online monitoring than the existing systems. Internet Iso-bar clusters hosts based on the similarity of their perceived network distance, and chooses the center of each cluster as a monitor site. The distance between two hosts is estimated using inter- or intra-cluster distances. Evaluation using real Internet measurements shows that Internet Iso-bar achieves high estimation accuracy and stability with much smaller measurement overhead than Global Network Positioning (GNP) [1]. Furthermore, by adjusting the number of clusters, we can smoothly trade off the measurement and management cost for better estimation accuracy.

[1]
T. S. E. Ng and H. Zhang, "Predicting Internet Network Distance with Coordinates-based Approaches," Proc. IEEE Infocom, New York, NY, June 2002.
1Microsoft Research

More information (http://www.cs.berkeley.edu/~yanchen) or

Send mail to the author : (yanchen@eecs.berkeley.edu)


Edit this abstract