University of Minnesota
University Relations
myU OneStop

Go to unit's home.

Home | Seminars and Symposia | Past seminars/symposia: Wednesday, March 20, 2002

DTC Seminar Series

Inference of IP Network Traffic Characteristics


Albert Greenberg
AT&T Labs

Wednesday, March 20, 2002
2:15 pm

402 Walter Library

OSPF (Open Shortest Path First) is a widely used intra-domain routing protocol in IP networks. Internal processing delays in OSPF implementations impact the speed at which updates propagate in the network, the load on individual routers, and the time needed for both intra-domain and inter-domain routing to reconverge following an internal topology or a configuration change. An OSPF user, such as an Internet Service Provider, typically has no access to the software implementation, and no way to estimate these delays directly. In this talk, we present black-box methods (i.e., measurements that rely only on external observations) for estimating and trending delays for key internal tasks in OSPF: processing Link State Advertisements (LSAs), performing Shortest Path First calculations, updating the Forwarding Information Base, and flooding LSAs. Corresponding measurements are reported for production routers from Cisco Systems. To help validate the methodology, black-box and white-box (i.e., measurements that rely on internal instrumentation) are reported for a open source (OSPF) distance calculations to break ties between candidates for traffic egress points. This is joint work with Aman Shaikh of UC Santa Cruz.


Dr. Greenberg received the B.A. degree in mathematics from Dartmouth College in 1978, and the M.S. and Ph.D. degrees in computer science from the University of Washington in Seattle in 1981 and 1983, respectively. He joined AT&T Bell Labs Mathematics Research Center in 1983, and became a Department Head in the Network Services Research Center in 1995. In 1996, he moved to AT&T Labs-Research, where he heads the IP Network Management and Performance Department, a group which conducts research in networking, with an emphasis on large scale IP networks and systems, and emerging Internet technologies. His research interests include Internet traffic measurement, modeling and engineering, policy based networking, and optical networking (IP/WDM). In collaboration with several others in AT&T Labs, he is developing a unified toolkit to manage IP networks. Dr. Greenberg's home page can be found at