WORKSHOP PROGRAM

All sessions will take place in the California Room unless otherwise noted.

Session papers are available to workshop registrants immediately and to everyone beginning April 27, 2010.

Programmable Networks and Their Applications

Automated and Scalable QoS Control for Network Convergence
Wonho Kim, Princeton University; Puneet Sharma, Jeongkeun Lee, Sujata Banerjee, Jean Tourrilhes, Sungju Lee, and Praveen Yalagandula, HP Labs

Read the Abstract | Full paper | Slides

The Case for Fine-Grained Traffic Engineering in Data Centers
Theophilus Benson, Ashok Anand, and Aditya Akella, University of Wisconsin—Madison; Ming Zhang, Microsoft Research

Read the Abstract | Full paper | Slides

HyperFlow: A Distributed Control Plane for OpenFlow
Amin Tootoonchian and Yashar Ganjali, University of Toronto

Read the Abstract | Full paper | Slides

Virtualization and Beyond

The "Platform as a Service" Model for Networking
Eric Keller and Jennifer Rexford, Princeton University

Read the Abstract | Full paper | Slides

vDC: Virtual Data Center Powered with AS Alliance for Enabling Cost-Effective Business Continuity and Coverage
Yuichiro Hei, KDDI R&D Laboratories, Inc.; Akihiro Nakao, The University of Tokyo; Tomohiko Ogishi and Toru Hasegawa, KDDI R&D Laboratories, Inc.; Shu Yamamoto, NICT

Read the Abstract | Full paper | Slides

Europa: Efficient User Mode Packet Forwarding in Network Virtualization
Yong Liao, University of Massachusetts at Amherst; Dong Yin, Northwestern Polytech University, China; Lixin Gao, University of Massachusetts at Amherst

Read the Abstract | Full paper | Slides

Measurement and Monitoring

A Preliminary Analysis of TCP Performance in an Enterprise Network
Boris Nechaev, Helsinki Institute for Information Technology; Mark Allman and Vern Paxson, International Computer Science Institute; Andrei Gurtov, Helsinki Institute for Information Technology

Read the Abstract | Full paper | Slides

Extensible and Scalable Network Monitoring Using OpenSAFE
Jeffrey R. Ballard, Ian Rae, and Aditya Akella, University of Wisconsin—Madison

Read the Abstract | Full paper | Slides

Beyond the Best: Real-Time Non-Invasive Collection of BGP Messages
Stefano Vissicchio, Luca Cittadini, Maurizio Pizzonia, Luca Vergantini, Valerio Mezzapesa, and Maria Luisa Papagni, Università degli Studi Roma Tre, Rome, Italy

Read the Abstract | Full paper | Slides

Network Management—Experiences

Experiences with Tracing Causality in Networked Services
Rodrigo Fonseca, Brown University; Michael J. Freedman, Princeton University; George Porter, University of California, San Diego

Read the Abstract | Full paper | Slides

Proactive Network Management of IPTV Networks
R.K. Sinha, K.K. Ramakrishnan, R. Doverspike, D. Xu, J. Pastor, A. Shaikh, S. Lee, and C. Chase, AT&T Labs—Research

Read the Abstract | Full paper | Slides

Panel

What Do Clouds Mean for Network and Services Management?

Moderator: Dr. Anees Shaikh, IBM T.J. Watson Research Center

Panelists: Adam Bechtel, Yahoo!; Tobias Ford, AT&T; Stephen Stuart, Google, Inc.; Changhoon Kim, Microsoft

The value of cloud computing in providing scalable infrastructure in a pay-as-you-go model has a strong appeal for enterprises looking to reduce the capital and operational expense of their IT environments. As the cloud is increasingly leveraged to deliver applications and services, new management opportunities and challenges arise. For example, clouds may be the ideal platform for providing scalable remote network management services, but perhaps at the cost of increased security and data privacy vulnerability, or with restricted functionality.

In this session, practitioners and technical leaders from industry will discuss some of the opportunities (and hype) around cloud-based management and areas where the research community should engage to be able to fully exploit the benefits of cloud computing. Some sample topics that will be discussed include:

• Does the cloud enable fundamentally new management service models?
• What are some of the new research challenges, for example related to problem and configuration management in the cloud?
• What are the primary barriers to adoption, or deployment, of cloud-based management services, for example security, latency, complexity?
• What approaches can help to overcome the cost and effort of migrating enterprise management to the cloud?

Panelist Bios
Adam Bechtel is the VP of Infrastructure Architecture and Strategy at Yahoo!. Prior to Yahoo!, Adam was the network architect for Inktomi. Prior to Inktomi, Adam was like everyone else working 100 hour weeks at a start-up. Adam has B.S. and M.S. engineering degrees from the University of Colorado.

As the Assistant VP of Technology for AT&T's Application Services business, Toby Ford leads technology efforts to enhance and streamline delivery of AT&T's offerings for cloud computing, hosting, outsourced application management, security, unified communications, telepresence, and intelligent content distribution service (ICDS). Previously, Mr. Ford held several positions in the U.S. and abroad including work with Cornell University, ARINC, TeleCommunication Systems, and his own company in the Netherlands. Toby holds a B.S. in Electrical Engineering from Cornell University.

Stephen Stuart is a Distinguished Engineer at Google, Inc., where he works on a variety of research problems in the networking space, including Software-Defined Networking. He is one of the tech leads for the Measurement Lab project, a joint effort of Google, New America Foundation, and PlanetLab, to improve network transparency in the Internet. Prior to joining Google, he built a global Internet backbone (AS6461), an exchange point (PAIX), Digital Equipment Corporation's west coast Internet presence (AS33, gatekeeper.dec.com, AltaVista), and wrote software for trading floor networks and CASE tools. He also serves on the board of the Mid-Peninsula Media Center, and helps to operate the F root nameserver.

Chang is an engineer in the Cloud Networking Team for Windows Azure and an affiliated member of Networking Research Group in Microsoft Research. He is primarily interested in building networking systems that get used by people. Since he joined Windows Azure, he has been working on cloud network virtualization with the goal of enabling cloud customers to build their own networks on a shared cloud. Chang received a Ph.D. from Princeton University where he worked on scalable and efficient self-configuring networks.