Rottscher Rottscher

Dr. Roger P. Karrer, PMP

Senior Research Scientist

Deutsche Telekom Laboratories
Ernst-Reuter-Platz 7
D-10587 Berlin
Germany

Tel: +49 30 8353 58459
Fax: +49 30 8353 58409
roger[dot]karrer[at]gmail[dot]com



General Information

Welcome to my personal research web site! I was a Senior Research Scientist at the Deutsche Telekom Laboratories in Berlin, Germany, from May 2005 to July 2008. Before, from November 2002 to March 2005, I was a PostDoc in the Rice Networks Group of Prof. Edward Knightly at Rice University. I got a PhD in Computer Science and an Engineering Diploma in Computer Science from ETH Zurich, Switzerland, in 2002. I worked there in the Lab for Software Technology at the Department of Computer Science . My PhD advisor was Prof. Thomas Gross. . After July 2008 I joined the Credit Suisse Group in Zurich, Switzerland.

I joined the Network Research Group of the Deutsche Telekom Laboratories (or brief T-Labs) because it offers a challenging and inspiring combination of academic freedom and applied, business-related research. In particular, I profit from the practical know-how of the Deutsche Telekom to deploy and run networks, the user-oriented objectives and the business- driven thinking. At the same time, I have the freedom and challenge to define my own research projects, to attract funding, to manage them and to successfully generate new knowledge of value for research in general and of interest for the Deutsche Telekom in particular.

News: Funding for Magnets and EU FP-7 projects on mesh networks and Clean Slate Internet Design

Magnets
The Magnets project that aims at deploying a high-speed WiFi mesh infrastructure, has received funding from the T-Labs for the 3rd year in a row. After deploying a high-speed WiFi backbone in 2005 and an indoor WiFi mesh network of 20 nodes in 2006, I received another grant to build an outdoor WiFi mesh network at the campus of the TU Berlin. The grant includes 100 high-speed mesh nodes and funding for a PhD student, system administrator, interns and undergrad students. With a size of 100 nodes and with each mesh node containing multiple WiFi cards, the testbed will be one of the most advanced WiFi mesh networks in the world. Moreover, the network will be integrated into the TU Berlin campus network and therefore serve as an operational network as well as a research testbed at the same time.
Currently, the Magnets testbed is under consideration for inclusion into national and European testbeds, such as G-Lab (project funded by the German government) and OneLab (EU).

EU FP-7 All three EU FP-7 project proposals I was involved in were selected for negotiations by the EU comission. These are:

  • Trilogy: an IP focusing on a clean slate design of resource control.
  • 4ward: an IP focusing on a clean slate Internet architecture, including e.g. Virtualization.
  • Carmen: a STREP focusing on carrier-grade mesh networks.
All the projects are still s.t. negotiations. The intended starting date is beginning of 2008.

3rd place at the German IT Security Award

My collegue Ulrich Kühn and I won the 3rd place of the First German Award for IT security of the Horst Görtz Stiftung for our work Edge-based Capabilities (EC). More information about the news is available from heise.de, the International School of IT Security (is-its), the Deutsche Telekom Website, or the Deutsche Telekom Laboratories Website.

Research interests, projects and activities

"There is nothing more certain and unchanging than uncertainty and change". This could not be more true for the Internet and its effects on our society. Over the past years we have created a network that we don't understand, that we don't control and that constantly changes itself and our everyday life.
My interests - maybe unexpectedly - are to contribute to the changes, either to make the Internet better or to understand it better. The questions that drive me are simple - and yet not that easy to answer (if you have an answer, do contact me asap!)
  • How much can we know about the underlying network - and how much knowledge is useful? The Internet is extremely complex - and the recent deployment of wireless access networks adds an additional dimension to it. I want to enhance our understanding of the network in terms of bandwidth, latency, topology from an experimental, application point of view. The Magnets project focuses on understanding heterogeneous wireless access (mesh) networks, and the Octopus project investigated how relevant network parameters can be measured and they can be used by distributed systems and applications in the (wired) Internet.
  • How do we lighten the dark side of the Internet - and how dark is it really? The Internet is fragile and vulnerable: outings, malware (worms, viruses), DDoS attacks, and (even) the fair resource usage. I want to find practical and easy to deploy solutions to reduce the fragility and vulnerability. The edge-based capabilities (EC) project follows these principles. In contrast, the TCP-Piranha project explores the dark side: how do I design the perfect DDoS attack? As of now, my intention is thereby not to actually launch it, but to be the first to find a counter-measure.
  • Do we need new network and application protocols? The Internet is evolving, and therefore new protocols are needed. New protocols can be adaptations or extensions of previous protocols or we may even look at a clean slate design. My objective is (i) to design new protocols and variants that address relevant new challenges, (ii) to evaluate them in a real network and (iii) to show their impact on real applications. Note here that all three points are a must - I am sick of all the deltaistic protocol proposals that are only evaluated in simulations! The Parallel download project with TCP-PARIS and other related protocols are an example of this.
So here is a brief overview of my (past and current) projects: More details can be found here

Professional activities

Publications

Access here

Disclaimer

This is my personal research Web page. The content of and any views expressed in this personal page are those of the person named only and are not those of the TU-Berlin, the Deutsche Telekom AG or the Deutsche Telekom Laboratories. The official Web page of the Deutsche Telekom Laboratories can be found here.