Network Support for Reliable Adaptive Distributed Systems

June 13

Seminar Introduction: The Current State of the Internet and the Case for Programmable Network Elements

"The Computer is the Network: The Emergence of Programmable Network Elements"
Advances in hardware are yielding a marriage of networking and processing in switched and routed infrastructures. Processing can now be embedded in the network, as "appliances" for many kinds of services. These provide protocol-aware classification of packets at line speeds, action invocation, and policy-based routing. Packet contents can be modified, delayed, or filtered based on user preferences, session context, or specified policies. A new generation of programmable network elements (PNEs) is becoming available, giving service and applications developers unprecedented control of processing inside the network. The migration of services into PNEs is the next leap in network service architecture. What is lacking is a unifying framework, able to span diverse programmable elements for different kinds of processing, and harnessing their capabilities in support of developers. We will describe an abstract specification model and supporting software suitable for unified programming of network services across diverse network elements. A focus is streaming video and storage, and the implementation of streaming video applications that exploit programmed services in the network, such as caches, managed buffers, and on-the-fly transcoders. PNEs can also form a pervasive monitoring capability at the network layer. This provides an extensive ability to observe-analyze-act on inferred network behaviors, enhancing the reliability and dependability of the network infrastructure underlying critical distributed applications.

June 14

Theme: What is the Structure of Internet Services and Why do They Fail?

  • Tutorial on system reliability: J. Gray, "Why do computers stop and what can be done about it?," Tandem TR 85.7, (June 1985). [PDF]
  • E. Brewer, "Lessons from Giant-Scale Services," IEEE Internet Computing, V 5., N. 4, (2001). [PDF] [Slides]
  • A. Fox, D. Patterson, "Self-Repairing Computers," Scientific American, (June 2003). [PDF]
  • D. Oppenheimer, A. Ganapathi, D. A. Patterson,"Why Do Internet Services Fail, and What Can Be Done About It?," USENIX Symp. On Internet Technologies and Systems, Seattle, WA, (March 2003). [PDF] [Slides]

June 15

Theme: Routing Anomalies and Instabilities

  • V. Paxson, "End-to-End Routing Behavior in the Internet," IEEE/ACM Transactions on Networking, Vol.5, No.5, pp. 601-615, October 1997. [Zipped PS]
  • C. Labovitz, G. R. Malan, F. Jahanian, “Internet Routing Instability,” IEEE/ACM Trans. On Networking, V. 6, N. 5, (October 1998), pp. 515-528. Original paper appeared in SIGCOMM 1997, Cannes, France, (August 1997). [PDF]
  • M. Roughan, T. G. Griffin, Z. M. Mao, A. Greenberg, B. Freeman, "IP Forwarding Anomalies and Improving their Detection Using Multiple Data Sources," Workshop on Network Troubleshooting: Research, Theory and Operations Practice Meet Malfunctioning Reality, ACM SIGCOMM, (August 2004), Portland, OR. [PDF]

June 16

Theme: Fires and Worms—Natural and Man-made Disasters and their Impact on Internet Reliability

  • Keynote Systems' White Paper [Doc] and Subsequent Press Releases [HTML, HTML, HTML]
  • J. Cowie, A. T. Ogielski, B. J. Premore, Y. Yuan, “Global Routing Instabilities Triggered by Code Red II and Nimda Worm Attacks,” Rensys Corporation White Paper, Hanover, NH, (December 2001). [HTML] [Slides]
  • L. Wang, Z. Zhao, D. Pei, R. Bush, D. Massey, A. Mankin, S. F. Wu, L. Zhang, “Observation and Analysis of BGP Behavior under Stress,” 2nd ACM Workshop on Internet Measurement, Marseille, France, (2002). [PDF] [Slides]

June 17

Theme: Active Networks, Middleboxes, and the Evolution of the Internet Architecture

  • D. L. Tennenhouse, J. M. Smith, W. D. Sincoskie, D. J. Wetherall, G. J. Minden, "A Survey of Active Network Research," IEEE Communications Magazine, V. 35, N. 1, (January 1997), pp. 80-86. [Zipped PS]
  • D. D. Clark, C. Partridge, J. C. Ramming, J. T. Wroclawski, "A Knowledge Plane for the Internet,"ACM SIGCOMM 2003, Karlsruhe, Germany, (August 2003). [PDF]
  • M. Walfish, J. Stribling, M. Krohn, H. Balakrishnan, R. Morris, S. Shenker, "Middleboxes No Longer Considered Harmful," USENIX OSDI 2004, San Francisco, CA, (December 2004). [PDF]
  • D. G. Andersen, H. Balakrishnan, M. F. Kaashoek, R. Morris, "Resilient Overlay Networks," 18th ACM SOSP, Banff, Canada, (October 2001), pp. 131-145. [PDF]