Welcome to the technical program of INM'06, presented on Monday 11 Sept 2006 as part of SIGCOMM'06.

The title of each paper is a link to the PDF of the published version of the paper.

Session 1: Changing the Network to be More Dynamic
09.00 - 10.30, Session Chair: Gisli Hjalmtysson

• RouterFarm - Towards a Dynamic, Manageable Network Edge|(slides)
  M. Agrawal (Carnegie Mellon University), S. R. Bailey (AT&T), A. Greenberg, J. Pastor, P. Sebos (AT&T Labs-Research), S. Seshan (Carnegie Mellon University), K. van der Merwe, J. Yates (AT&T Labs-Research)
• Building a Controlled Delay Assured Forwarding Class in Differentiated Services Networks|(slides)
  P. Kulkarni, M. Nazeeruddin, S. McClean (University of Ulster, Coleraine)
• An Integrated, Distributed Traffic Control Strategy for the Future Internet|(slides)
  H. Che (University of Texas at Arlington), W. Sub, C. Lagoab (Pennsylvania State University), K. Xuc, C. Liuc, Y. Cuic (Tsinghau University, Beijing)
• A Virtual Disk based Centralized Management Paradigm for Enterprise Networks|(slides)|(project web page)
  Y. Zhou, Y. Zhang (Tsinghua University), Y. Xie (Carnegie Mellon University)

Session 2: Changing the Architecture of Network Management
11.00 - 12.30, Session Chair: Geoffrey Xie

• Dynamic Connectivity Management with an Intelligent Route Service Control Point|(slides)
  J. Van der Merwe, A. Cepleanu, K. D’Souza, B. Freeman, A. Greenberg, D. Knight, R. McMillan, D. Moloney, J. Mulligan, H. Nguyen, M. Nguyen, A. Ramarajan, S. Saad, M. Satterlee, T. Spencer, D. Toll, S. Zelingher (AT&T Labs)
• On Information Hiding and Network Management|(slides)
  K. L. Calvert, J. Griffioen (University of Kentucky)
• CONMan - Taking the Complexity out of Network Management|(slides)
  H. Ballani, P. Francis (Cornell University, Ithaca)
• Towards a Framework for Network Control Composition|(slides)
  T. S. Eugene Ng (Rice University), H. Yan (Carnegie Mellon University)

Session 3: Measurement for Management
13.30 - 15.00, Session Chair: Eugene Ng

• A Measurement-Friendly Network (MFN) Architecture
  S. Machiraju (Sprint ATL), D. Veitch (University of Melbourne)
• VMScope - A Virtual Multicast VPN Performance Monitor
  L. Breslau (AT&T Research), C. Chasey (SBC Labs), N. Duffield, B. Fenner (AT&T Research), Y. Mao (University of California at San Diego), S. Sen (AT&T Research)
• Adaptive Distributed Monitoring with Accuracy Objectives|(slides)
  A. Gonzalez Prieto, R. Stadler (KTH Royal Institute of Technology)
• S3: A Scalable Sensing Service for Monitoring Large Networked Systems|(slides)|(project website)
  P. Yalagandula, P. Sharma, S. Banerjee, S. Basu, S.-J. Lee (HP Labs, Palo Alto) [The project website has data from our deployment on Planetlab available to the public.]

Session 4: Policy and Management
15.30 - 16.45,Session Chair: Rolf Stadler

• Policy-based BGP Control Architecture for Autonomous Routing Management|(slides)
  O. Akashi (NTT Network Innovation Labs), K. Fukuda (National Institute of Informatics), T. Hirotsu (Toyohashi University of Technology/JST CREST), T. Sugawara (NTT Communication Science Labs)
• Optimal Strategy for Graceful Network Upgrade|(slides)
  R. Keralapura, C.-N. Chuah, Y. Fan (University of California at Davis)
• Autonomic Network Management - Some Pragmatic Considerations|(slides)
  R. Mortier, E. Kiciman (Microsoft Research)

Panel: Can we do experimental research in network management?
17.00 - 18.00, Panel Organizers: Paul Francis and Taieb Znati

Panelists:
 Kenjiro Cho (IIJ and WIDE)
 Jay Lepreau (University of Utah and Emulab)
 Jennifer Rexford (Princeton University)

While researchers have had success in studying specific aspects of network management, such as routing algorithms and anomaly detection, there has been little success in the study of network management "in the large": that is, the broader problems of planning, installation, configuration, provisioning, testing, monitoring, and debugging. No doubt one of the reasons is that network management in the large is such a vast problem space. Another reason, however, is that we have few if any good tools and methods for doing experimental network management. This panel explores issues in experimental network management research: Do we need a network management data repository? If so, what should be in that repository and how can we obtain it? Can we build synergy towards an open set of network management tools, and if so how can those tools be used in practice? How can we exploit emerging experimental research infrastructures like GENI? For instance, how can different management experiments be run on GENI virtual networks without compromising those networks? These and other questions will be discussed.

Questions posed to the panel.