• home
• current students
• graduate
• events

Presenter: Bernhard Plattner - Computer Engineering and Networks Laboratory, ETH, Zurich, Switzerland
Supervisor:

Date: Fri, September 6, 2002
Time: 13:30:00 - 00:00:00
Place: EOW 430

ABSTRACT

Abstract: In this talk, I will start out with a brief review of the roots of active networking research and in the following attempt to characterize the state of the art in active networking research, as it presents itself after five years of research into the topic. The main purpose of the talk will be, however, to identify important areas for future research around active networking. Active networking is not really a new idea. Back in the mid-eighties already, research in what we call today active networking has been undertaken, e.g. at KTH in Stockholm, Sweden, with the SOFTNET project. This project aimed at providing a playground for research in radio-based mobile systems and used a virtual machine based approach with programs written in FORTH as capsules. Active networking was re-discovered in 1995 by Tennenhouse and Wetherall. Their publications spurred research in many places, and eventually led to the creation of the DARPA research program on active networks. The DARPA program produced a basic understanding about the design space which can be used by designers of active routers and applications. The third phase in AN research (after the early experiments and the DARPA program) is now in progress. It is characterized by the recognition that the "capsules" approach, which has long been equated with active networking, is not sufficient to guarantee the security and safety requirements the networking community has. It is apparent that research has shifted towards combining ideas developed in the DARPA research with the discipline of programmable networks, which actually means that the two camps, both aiming at creating a flexible future network architecture, are merging. The challenge will be to create frameworks for creating and deploying services in a heterogeneous environment, with multiple execution environments built by different manufacturers. It is obvious that defining and deploying such frameworks will only be possible if some standardization can be achieved (this, in principle, dismisses one of the promises of the active networking community - to provide a new network architecture without the need for extensive, time consuming and costly standardization). Active networks research is now at midway. To create the frameworks mentioned previously is one important step ahead; however, another, equally important one is to involve industry (manufacturers and service providers) and - last but not least - users. This talk therefore closes with a call to assure that research and development can proceed by opening the ways for transferring the basic technology developed so far to these new communities.

Biography:

Dr. Plattner is a Professor of Computer Engineering at ETH Zurich, where he leads the communication systems research group. His research focuses on exploring new approaches to protocol and network engineering, such as active and programmable networks. He is also interested in the development of multimedia applications for high-speed networks. Dr. Plattner is a member of the ieee, acm and the Internet Society. He served as the program chair of various important conferences, such as ACM SIGCOMM 91 and INET '94. He also served as a Vice President of TERENA, the European association of research and education networks.