On the stability of network distance estimation

Overlay network distance monitoring and estimation system can benefit many new applications and services, such as peer-to-peer overlay routing and location. However, there is a lack of such scalable system with small overhead, good usability, and good distance estimation accuracy and stability. Thus we propose a scalable overlay distance monitoring system, Internet Iso-bar, which clusters hosts based on the similarity of their perceived network distance, with no assumption about the underlying network topology. The centers of each cluster are then chosen as monitors to represent their clusters for probing and distance estimation. We compare it with other network distance estimation systems, such as Global Network Positioning (GNP) [1]. Internet Iso-bar is easy to implement and use, and has good scalability and small communication and computation cost for online monitoring. Preliminary evaluation on real Internet measurement data also shows that Internet Iso-bar has high prediction accuracy and stability. Finally, by adjusting the number of clusters, we can smoothly trade off the measurement and management cost for better distance estimation accuracy and stability.

[1]  Lixia Zhang,et al.  On the placement of Internet instrumentation , 2000, Proceedings IEEE INFOCOM 2000. Conference on Computer Communications. Nineteenth Annual Joint Conference of the IEEE Computer and Communications Societies (Cat. No.00CH37064).

[2]  Frank Thomson Leighton.  The Challenges of Delivering Content and Applications on the Internet , 2004, RIDE.

[3]  Sudipto Guha,et al.  Improved combinatorial algorithms for the facility location and k-median problems , 1999, 40th Annual Symposium on Foundations of Computer Science (Cat. No.99CB37039).

[4]  Balachander Krishnamurthy,et al.  On network-aware clustering of Web clients , 2000, SIGCOMM 2000.

[5]  Yin Zhang,et al.  On the constancy of internet path properties , 2001, IMW '01.

[6]  Hui Zhang,et al.  Predicting Internet network distance with coordinates-based approaches , 2002, Proceedings.Twenty-First Annual Joint Conference of the IEEE Computer and Communications Societies.

[7]  T. Grossman,et al.  Computational Experience with Approximation Algorithms for the Set Covering Problem , 1994 .

[8]  Kurt Rothermel,et al.  Dynamic distance maps of the Internet , 2000, Proceedings IEEE INFOCOM 2000. Conference on Computer Communications. Nineteenth Annual Joint Conference of the IEEE Computer and Communications Societies (Cat. No.00CH37064).

[9]  Mark Handley,et al.  Topologically-aware overlay construction and server selection , 2002, Proceedings.Twenty-First Annual Joint Conference of the IEEE Computer and Communications Societies.

[10]  Paul Francis,et al.  IDMaps: a global internet host distance estimation service , 2001, TNET.

[11]  Ibrahim Matta,et al.  On the origin of power laws in Internet topologies , 2000, CCRV.

[12]  Hari Balakrishnan,et al.  Resilient overlay networks , 2001, SOSP.

[13]  S. Jamin.  On the placement of internet instrumentaion , 2000 .

[14]  Balachander Krishnamurthy,et al.  On network-aware clustering of Web clients , 2000, SIGCOMM.

[15]  Paul Francis,et al.  An architecture for a global Internet host distance estimation service , 1999, IEEE INFOCOM '99. Conference on Computer Communications. Proceedings. Eighteenth Annual Joint Conference of the IEEE Computer and Communications Societies. The Future is Now (Cat. No.99CH36320).