Workshop on
Future Directions in Network Architecture (FDNA-03)

Karlsruhe, Germany, August 27, 2003
In conjunction with ACM SIGCOMM 2003

Call For Papers

The architecture of a network specifies the high level principles and structures that guide the network's design, especially the engineering of its protocols and algorithms, and the interaction of its different functional components. Architectural development operates at two distinct layers:

  • Identification of fundamental structuring principles
  • Decisions about functional decomposition and system modularity

The current Internet architecture has been remarkably successful as the underpinning of a global, general-purpose, decentralized data communication network. Architectural decisions made 30 years ago have allowed the Internet to quickly support new applications and adapt itself to dramatic changes in technology. Now, new forces arise.

  • New classes of networks - sensor-nets, highly mobile ad-hoc nets, overlays, and others - have come into existence. These networks have very different design goals, operating requirements, and implementation environments than those imagined for traditional network architectures.
  • The requirements underlying the architecture of general-purpose networks such as the Internet have changed significantly since the 1970's. The coherence of the Internet design is being eroded by a patchwork of narrow technical embellishments to satisfy changing requirements. The result is greatly increased complexity and some loss of functionality and extensibility. Revisiting the architectural principles of large general-purpose networks may be appropriate at this point.

This call solicits papers on the broad topic of new developments in network architecture. Examples of topics include:

  • Architec tural principles for complex, decentralized networks
  • Self-sustaining architecture (architecture that discourages violation of its core principles)
  • Relationship of technical architecture to social and economic issues
  • Architectural aspects of overlay and multi-layer networks
  • Architecture of robust and trustworthy networks
  • Principles and architecture of highly scalable systems
  • Architectures for data-centric networks (e.g., sensornets and related concepts)

Submissions ranging from presentations of specific research to more general, philosophical position papers are welcome. Papers that bring out interesting and novel ideas at an early stage in their development are favored over highly polished, journal-style results. Selected papers will be forward-looking, with impact and implications for ongoing or future research.

Papers submitted in response to this call will be considered both for discussion at the Sigcomm 2003 Workshop on Future Directions in Network Architecture, and for publication in a special section of Computer Communications Review. Submitted papers may be accepted for either or both of these forums. Papers will be considered for both forums by a unified review committee.

Papers may be dual-submitted to the Sigcomm 2003 Conference and this call for papers. Dual-submitted papers must be completed by the Sigcomm conference deadline, and must be clearly marked as dual submissions. These papers will be reviewed by both PC's, and may be accepted by one or both events. Papers accepted to both events will appear only in the Sigcomm conference proceedings.

Paper Format and Submission Instructions

Please see the FDNA paper submission guidelines.

Important Dates

Submission deadline March 17, 2003
Notification of acceptance May 27, 2003
Final version papers due June 12, 2003
Workshop date August 27, 2003
  • Submission deadline extensions of up to one week may be requested by email to fdna@lcs.mit.edu. All such requests must be received and acknowledged by the March 17th deadline. A paper title, abstract, and author list must be submitted by the March 17th deadline even if an extension is granted.
  • Please note that the final paper due date, June 12, is earlier than that published in previous versions of the CFP. This is due to Sigcomm's decision to print a unified volume of workshop papers, to be available at the conference.

Organizing Committee

  • Andrew Campbell, Columbia
  • Ted Faber, USC ISI
  • Mark Handley, ICRI
  • John Heidemann, USC ISI
  • Larry Peterson, Princeton
  • James Sterbenz, BBN
  • John Wroclawski, MIT (Chair)

Further Information

For further information about FDNA-03 is available at http://www.acm.org/sigcomm/FDNA-03. Alternatively, please contact the organizing committee by email at fdna@lcs.mit.edu.