1 / 18

NSF Workshop on Fundamental Research in Networking: Future Directions and Visions for Networking Research

NSF Workshop on Fundamental Research in Networking: Future Directions and Visions for Networking Research Ness B. Shroff School of Electrical and Computer Engineering Network Engineering and Wireless Systems (NEWS) Group Purdue University, West Lafayette, IN 47907 NSF Workshop

Télécharger la présentation

NSF Workshop on Fundamental Research in Networking: Future Directions and Visions for Networking Research

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.


Presentation Transcript

  1. NSF Workshop on Fundamental Research in Networking: Future Directions and Visions for Networking Research Ness B. ShroffSchool of Electrical and Computer EngineeringNetwork Engineering and Wireless Systems (NEWS) GroupPurdue University, West Lafayette, IN

  2. NSF Workshop • The NSF Workshop on Fundamental Research in Networkingwas held in Arlie House, Virginia, April 24-25. • In attendance: 27 Experts spanning a broad spectrum of networking areas and disciplines. Mission: Prepare a report on the current state of networking and to articulate a vision for the future of fundamental research in networking

  3. NSF Workshop: Logistics and Ground Rules • Two day workshop that was mission focused • Only 2 panel sessions • A lot of time in breakout groups • Writing began during the workshop • Not to simply have an intellectual discussion • Not to write a report consisting of a laundry list of people's favorite topics • Represent the community at large, • Hope that report will stimulate far-reaching future research initiatives and collaborations.

  4. Fundamental Research Fundamental Research was defined as having one or more characteristics of being far-reaching, solid, long-term, paradigm-changing, risk-taking, disruptive, high-impact research that could be funded by a program such as the Network Research program at NSF. • Spans the entire spectrum of networking research, from theory to systems-oriented research

  5. Executive Summary: Findings and Recommendations Models of Success Case studies of networking research successes Network Innovator’s dilemma Need for reproducibility and benchmarking Importance and Impact of Basic Research in Networking New Applications and Paradigms Grand Challenges in Networking Quick Look at Workshop Report Focus on selected items from report here … • Full report is availableat this conference and also on

  6. Network Innovator’s Dilemma • Looking to models of success for future research, we must avoid the temptation of falling into the trap of Innovator’s Dilemma (Clayton Christenson) • For example: • Success of the Internet carries the risk of complacency about the need for out-of-the-box thinking. • Fail to observe disruptive technologies that take radical new shape • Past: Needed new community to think radically (e.g., bell-heads were left behind in the Internet boom) • Future: Do it ourselves!

  7. Importance of Basic Research • Basic Research in Networking addresses more foundational and theoretical aspects of networking. • Brings together methodologies from a variety of different fields to address specific issues of relevance to networking. • Aims: • develop insights and principles • understand performance limitations • suggest “good” solutions and policies • guide the development of protocols and control mechanisms

  8. Importance of Basic Research • Basic research: key role in the understanding and development of Internet and wireless networks • Distributed routing protocols (Bellman-Ford and Dijkstra). • Multiaccess mechanisms from ALOHA to CSMA/CD • TCP and more general AIMD congestion control • 3G scheduling mechanisms in wireless systems • Queueing and Large Deviations lead to understanding of network congestion and resource allocation • LRD properties in network traffic… • Also provided the impetus for advancements in associated fields. • Needed now more than ever to design and control networks on an even grander scale than before! • Greater impact if basic research is pursued in conjunction with focus areas and real problems.

  9. New Applications and Paradigms • Applications are and will continue to be the raison d’etre of networks! • 3 illustrative applications were chosen as likely to play an important role driving future networking. • Ubiquitous Tele-presence (ability to interact with others and environment without shackles of space) • Universal Sensing (large-scale sensor networks integrated into all aspects of life). • Emergency Response and Disaster Recovery Networks • Applications will induce user focused as well as network centric challenges • Question: Are there a set of grand challenges for networking?

  10. Grand Challenges Classical definition "Grand Challenges are fundamental problems in science and engineering with potentially broad social, political, economic, and scientific impact that can be advanced by applying high performance computing resources.“ Definition used in Workshop "As a fundamental problem with broad economic and scientific impact that is currently considered intractable, and whose solution can be advanced by fundamental research."

  11. Grand Challenges Meta Grand Challenge To develop new network theories, architectures, and methodologies that will facilitate the development of the next generation of services and applications • Indvidual Grand Challenges • Network Information Theory for Wireless Networks • Providing Market Incentives (The Economics of Networking) • Development of resilient Networks • Sensorized Universe • Design of Overlay Networks • Virtual Networks Discuss only first three bullets

  12. Network Information Theory for Wireless Networks • Classical Information theory: basis for major developments in communications, error control, and signal processing over the last 50 years. • Main limitation: characterization of delay is absent from traditional information theory. • Many important theoretical developments (scheduling, queueing, control, routing), there is no such unified information theory for networks.

  13. Network Information Theory • Challenges: • Characterize the fundamental limits performance limites of mobile and static wireless networks • Capacity • Delay • Power • Connectivity and Coverage • Can we develop distributed algorithms to achieve these limits? • Network Coding? • Can we achieve orders of magnitude increases in network peformance by exploiting underlying technologies (physical layer, VLSI design, etc.)?

  14. The Economics of Networking • Lack of Marketing Incentives are one of the main impediments to the deployment of new services on the Internet. • Recent history has demonstrated that the bottleneck to improving the global network may be economics and not technology. • There has been enormous interest in pricing to manage networks. • However, most works do not consider a competitive market environment. • But the Internet is not a single service provider!

  15. The Economics of Networking • Major Challenge: In developing economic mechanisms for the creationof efficient markets: • Allow everyone to be a player (not simply social maximization) • Designed for multi-provider and multi-business environment (heterogeneous) • Incentive-compatible (Don’t assume altruism) • Scalable (12,000+ AS!) • Flexible: Support different service types (e.g., Dependable service, variable price; Dependable price, variable service) Increases Revenue Quality Improves

  16. Resilient Networking • As the Internet grows in scale it faces an increasingly adverse environment • Due to the sheer scale, faults become quite commonplace, rather than the exceptions • Component failure • Human operational errors • Spread of software viruses • Malicious attacks.

  17. Resilient Networking • Challenge:To provide the Internet with a resilientinfrastructure. • Need to develop a basic understanding of the rules and principles in network protocol design that facilitate the realization of resiliency in large-scale networks. • Can we develop specific enhancements that can readily be applied to the current Internet to effectively strengthen its defense ability? • Can we really prepare against the unexpected?

  18. Conclusion • Networking has seen unprecedented success and growth • Future is bright and holds even greater and grander challenges • Our mission: Innovate, conduct high-quality research, and educate industry and government agencies of the enormous continued importance of the field. Thank You!

More Related