Check out the new USENIX Web site.
USENIX '06 Home  |   Registration  |   Organizers  |   Invitation  |   At a Glance  |  Training  |  Tech Sessions
FAST-OS Workshop  |   Poster Session  |   BoFs  |  Sponsors  |  Activities  |  Hotel/Travel  |  Services  |  Students
Questions?  |  Help Promote!  |  Brochure PDF  |  Call for Papers  |  Past Proceedings  |  Authors  |  Speakers

Technical Sessions: Thursday, June 1 | Friday, June 2 | Saturday, June 3
Thursday, June 1
8:45 a.m.–10:30 a.m. Thursday
Opening Remarks, Awards, Keynote

Keynote Address
PlanetLab: Evolution vs. Intelligent Design in Planetary-Scale Infrastructure
Larry Peterson, Professor and Chair, Department of Computer Science, Princeton University; Director, PlanetLab Consortium

PlanetLab is a global platform for evaluating and deploying network services. It currently includes over 600 nodes, spanning nearly 300 sites and 30 countries, and hosts over 400 experimental services. PlanetLab must satisfy a unique set of sometimes contradictory requirements: based on our experiences building PlanetLab over the past three years, we are now able to define an architecture that satisfies these requirements. This talk identifies the requirements, presents the design principles that follow from them, and outlines the resulting PlanetLab architecture. It also briefly discusses some of the lessons we learned about building large network systems.

10:30 a.m.–11:00 a.m.   Break  
11:00 a.m.–12:30 p.m. Thursday
SYSTEMS PRACTICE & EXPERIENCE REFEREED PAPERS TRACK

Virtualization

Antfarm: Tracking Processes in a Virtual Machine Environment
Stephen T. Jones, Andrea C. Arpaci-Dusseau, and Remzi H. Arpaci-Dusseau, University of Wisconsin, Madison

Optimizing Network Virtualization in Xen
Aravind Menon, EPFL; Alan L. Cox, Rice University; Willy Zwaenepoel, EPFL

High Performance VMM-Bypass I/O in Virtual Machines
Jiuxing Liu, IBM T.J. Watson Research Center; Wei Huang, The Ohio State University; Bulent Abali, IBM T. J. Watson Research Center; Dhabaleswar K. Panda, The Ohio State University

INVITED TALKS

Deploying a Sensor Network on an Active Volcano
Matt Welsh, Harvard University

I will describe our experience deploying a wireless sensor network on Reventador, an active volcano in Ecuador, which recorded data on hundreds of eruptions and explosions over a three-week period. Deploying a sensor network in such a hostile and remote environment involves many challenges, including capturing high-quality data on seismic activity, debugging the network over the radio, and slogging through jungles to reach the deployment site.

THE GURU IS IN

Computers, Developers, and The Law
Dan Appelman, Heller Ehrman, LLP

Dan is a partner in a major law firm in the Silicon Valley and is legal counsel for USENIX. He advises clients about all aspects of law relating to the development and commercialization of technology, including intellectual property rights, privacy, obligations to one's employer, issues when selling (or buying) a technology company, issues arising from use of open source code, and raising money from investors.

This session will be a free-wheeling, anything goes get-together where you can ask the questions that have always bugged you about your favorite legal issues. Dan will give you the current status of the hottest legal issues related to technology and those who develop technology. Whether your concern is how to avoid entanglement with the federal government under the U.S. Patriot Act or how to protect the privacy of your personal information on the web, Dan will tell you what your legal rights are and will give you practical advice. Until questions overtake him and completely overwhelm his agenda, Dan will start off with a discussion of common pitfalls to avoid in dealing with employers when you are independently developing code.

12:30 p.m.–2:00 p.m.   Lunch   (on your own)  
2:00 p.m.–3:30 p.m. Thursday
SYSTEMS PRACTICE & EXPERIENCE REFEREED PAPERS TRACK

Storage

Provenance-Aware Storage Systems
Kiran-Kumar Muniswamy-Reddy, David A. Holland, Uri Braun, and Margo Seltzer, Harvard University

Thresher: An Efficient Storage Manager for Copy-on-write Snapshots
Liuba Shrira and Hao Xu, Brandeis University

Design Tradeoffs in Applying Content Addressable Storage to Enterprise-scale Systems Based on Virtual Machines
Partho Nath, Penn State University; Michael A. Kozuch, Intel Research Pittsburgh; David R. O'Hallaron, Jan Harkes, M. Satyanarayanan, Niraj Tolia, and Matt Toups, Carnegie Mellon University

INVITED TALKS

Panel: Open Source Software Business Models
Moderator: Stephen Walli,Optaros, Inc.

Panelists: Mike Olsen, Oracle/Sleepycat; Brian Aker, MySQL; Miguel de Icaza, Novell/Ximian

The panelists come from a variety of backgrounds and software companies. The one thing they all have in common is that they have used free and open source software in those businesses to great effect. Come find out how the business models worked, how community made a difference, and just what an open source business looks like.

THE GURU IS IN

Hacking Perl²
David N. Blank-Edelman, Northeastern University CCIS

David is the Director of Technology at the Northeastern University College of Computer and Information Science and the author of the O'Reilly book Perl for System Administration. He has spent the last 20+ years as a system/network administrator in large multi-platform environments. He's been a Perl user for more than 10 years.

In this session we'll be talking about the process of hacking Perl and the Perl hacks that can make working in the language easier. If you have a problem that you think Perl code may be able to address, stop by and he'll try to get you pointed in the right direction. Given David's background, expect a strong sysadmin bent to the session, but you are welcome to bring questions from any field and we'll give them a whirl.

3:30 p.m.–4:00 p.m.   Break  
4:00 p.m.–5:30 p.m. Thursday
SYSTEMS PRACTICE & EXPERIENCE REFEREED PAPERS TRACK

Short Papers Session I

Compare-by-Hash: A Reasoned Analysis
J. Black, University of Colorado, Boulder

An Evaluation of Network Stack Parallelization Strategies in Modern Operating Systems
Paul Willmann, Scott Rixner, and Alan L. Cox, Rice University

Disk Drive Level Workload Characterization
Alma Riska and Erik Riedel, Seagate Research

Towards a Resilient Operating System for Wireless Sensor Networks
Hyoseung Kim and Hojung Cha, Yonsei University

Transparent Contribution of Memory
James Cipar, Mark D. Corner, and Emery D. Berger, University of Massachusetts, Amherst

INVITED TALKS

Success, Failure, and Alternative Solutions for Network Security
Peiter "Mudge" Zatko, BBN Technologies

The computer and network security fields have made little progress in the past decade. Why have our responses not solved our problems? The explanation that attacks are becoming ever more complicated and thus defenses are continually behind makes little sense when 10-year-old rootkits and well-known BGP and DNS attacks are still in force. Yes, it is still possible to take down the entire Internet.

This talk looks at the networked environment and its growth and at the inadequate responses of the Internet security industry. It then presents corollaries for environments in different academic disciplines and examines certain psychological situations that prohibit researchers and organizations from being able to address the problems. Finally, it suggests counterintelligence and counterespionage models and applies them to low-level network and systems communications.

THE GURU IS IN

Algorithms and Data Structures for the 21st Century
Stephen C. Johnson, The MathWorks, Inc.

Today's (and tomorrow's) computers are very different from those of a few decades ago, and are becoming ever more different. Data structures and algorithms that were state of the art in the past may be suboptimal or even unusable on future machines. I have no answers, but I do have some data and some speculation, and invite you bring your experiences and speculate with me.

Steve Johnson is a former president of USENIX who has been involved with C and UNIX since their earliest days. He is the author of Yacc, lint, and the Portable C Compiler. He also spent 15 years in Silicon Valley, most recently at Transmeta. He is currently employed at The MathWorks, Natick, MA, where he works on the MATLAB language.

6:00 p.m.–7:30 p.m. Thursday
Poster Session & Happy Hour

Don't miss the cool new ideas and the latest preliminary research on display at the Poster Session & Happy Hour. Take part in discussions with your colleagues over complimentary drinks and hors d'oeuvres.
Technical Sessions: Thursday, June 1 | Friday, June 2 | Saturday, June 3
Friday, June 2
9:00 a.m.–10:30 a.m. Friday
SYSTEMS PRACTICE & EXPERIENCE REFEREED PAPERS TRACK

Server Implementation

Implementation and Evaluation of Moderate Parallelism in the BIND9 DNS Server
Tatuya Jinmei, Toshiba Corporation; Paul Vixie, Internet Systems Consortium

Flux: A Language for Programming High-Performance Servers
Brendan Burns, Kevin Grimaldi, Alexander Kostadinov, Emery D. Berger, and Mark D. Corner, University of Massachusetts, Amherst

Understanding and Addressing Blocking-Induced Network Server Latency
Yaoping Ruan, IBM T.J. Watson Research Center; Vivek Pai, Princeton University

INVITED TALKS

Panel: Is University Systems Teaching and Research Relevant to Industry?
Moderator: Gernot Heiser, NICTA/UNSW

This panel will discuss the question of whether systems research published in the main conferences is too academic, and whether universities are turning out enough graduates with the right systems skills for industry. It will also look at whether industry could be more supportive of university teachers and researchers.

THE GURU IS IN

Sure, You Can Archive Data, But Will You Be Able to Retrieve It in 20 Years?
Evan Marcus, Deputy CTO, Archivas Software

Every business has fixed content data that must be safely stored for the long-term. Whether it's medical records, corporate financial data, security data, old photographs, or an MP3 collection, the data must be preserved.

But, will you be able to get it back when you need it? If you cannot, then you shouldn't have bothered storing it in the first place. In this session, we'll look at the three key functions that any data archive must perform: ingestion, preservation, and retrieval, and we'll look at some of the current solutions on the market, and how they perform this vital function.

Evan Marcus is the Deputy CTO at Archivas, Inc. The Archivas Cluster is an open-hardware cluster, designed for the long-term archiving and retrieval of fixed-content data.

10:30 a.m.–11:00 a.m.   Break  
11:00 a.m.–12:30 p.m. Friday
SYSTEMS PRACTICE & EXPERIENCE REFEREED PAPERS TRACK

Security

Reval: A Tool for Real-time Evaluation of DDoS Mitigation Strategies
Rangarajan Vasudevan and Z. Morley Mao, University of Michigan; Oliver Spatscheck and Jacobus van der Merwe, AT&T Labs—Research

LADS: Large-scale Automated DDoS Detection System
Vyas Sekar, Carnegie Mellon University; Nick Duffield, Oliver Spatscheck, and Jacobus van der Merwe, AT&T Labs—Research; Hui Zhang, Carnegie Mellon University

Bump in the Ether: A Framework for Securing Sensitive User Input
Jonathan M. McCune, Adrian Perrig, and Michael K. Reiter, Carnegie Mellon University

INVITED TALKS

Architectures and Algorithms for Biomolecular Simulation
Cliff Young, D.E. Shaw Research, LLC

Many important outstanding questions in the fields of biology, chemistry, and medicine could in principle be answered through the atomic-level simulation of biologically significant molecules. Current technology and algorithms, however, fall several orders of magnitude short of the power that would be required to do so. This talk describes the state of the art in biomolecular simulation and explores the potential role of high-performance computing technologies in extending current capabilities. Efforts within our own lab to develop novel architectures and algorithms for the acceleration of molecular dynamics simulations by several orders of magnitude will be described.

THE GURU IS IN

RRDTool: Logging and Graphing
John Sellens, SYONEX

RRDtool is one of the basic building blocks in a system or network administrator's toolkit, and is used standalone and in numerous applications for storing and graphing "time series data." This session will introduce RRDtool, how it works, and some of its applications.

John Sellens is a system administrator and USENIX tutorial instructor who thinks that RRDtool is one of the coolest things.

12:30 p.m.–2:00 p.m.   Lunch   (on your own)  
2:00 p.m.–3:30 p.m. Friday
SYSTEMS PRACTICE & EXPERIENCE REFEREED PAPERS TRACK

Management and Administration

Sharing Networked Resources with Brokered Leases
David Irwin, Jeff Chase, Laura Grit, Aydan Yumerefendi, and David Becker, Duke University; Kenneth G. Yocum, University of California, San Diego

Understanding and Validating Database System Administration
Fábio Oliveira, Kiran Nagaraja, Rekha Bachwani, Ricardo Bianchini, Richard P. Martin, and Thu D. Nguyen, Rutgers University

SMART: An Integrated Multi-Action Advisor for Storage Systems
Li Yin, University of California, Berkeley; Sandeep Uttamchandani, Madhukar Korupolu, and Kaladhar Voruganti, IBM Almaden Research Center; Randy Katz, University of California, Berkeley

INVITED TALKS

Permissive Action Links, Nuclear Weapons, and the History of Public Key Cryptography
Steven M. Bellovin, Columbia University

From a security perspective, command and control of nuclear weapons presents a challenge. The security mechanisms are supposed to be so good that they're impossible to bypass. But how do they work? Beyond that, there are reports linking these mechanisms to the early history of public key cryptography. We'll explore the documented history of both fields and speculate on just how permissive action links—the "combination locks" on nuclear weapons—actually work.

THE GURU IS IN

Security Is Broken
Rik Farrow, Security Consultant

Our computer security model is broken. In fact, it never really has worked well, but is even less suitable for today's users. In this session, I will explain why I feel that current security software and OS design are nothing more than bandaids, and why a totally new way of thinking about computer security is mandatory. Of course, I will be willing to answer more general computer security questions, but will lead off with "Security is broken." Plan on participating in a lively discussion.

Rik Farrow provides UNIX and Internet security consulting and training. He has been working with UNIX system security since 1984 and with TCP/IP networks since 1988. He has taught at the IRS, Department of Justice, NSA, NASA, US West, Canadian RCMP, Swedish Navy, and for many U.S. and European user groups. He is the author of UNIX System Security, published by Addison-Wesley in 1991, and System Administrator's Guide to System V (Prentice Hall, 1989). Farrow is the editor of ;login: and a network security columnist for Network magazine. Rik lives with his family in the high desert of northern Arizona and enjoys hiking and mountain biking when time permits.

3:30 p.m.–4:00 p.m.   Break  
4:00 p.m.–5:15 p.m. Friday
SYSTEMS PRACTICE & EXPERIENCE REFEREED PAPERS TRACK

Short Papers Session II

sMonitor: A Non-Intrusive Client-Perceived End-to-End Performance Monitor of Secured Internet Services
Jianbin Wei and Cheng-Zhong Xu, Wayne State University

Privacy Analysis for Data Sharing in *nix Systems
Aameek Singh, Ling Liu, and Mustaque Ahamad, Georgia Institute of Technology

Securing Web Service by Automatic Robot Detection
KyoungSoo Park and Vivek S. Pai, Princeton University; Kang-Won Lee and Seraphin Calo, IBM T.J. Watson Research Center

Cutting through the Confusion: A Measurement Study of Homograph Attacks
Tobias Holgers, David E. Watson, and Steven D. Gribble, University of Washington

Stealth Probing: Efficient Data-Plane Security for IP Routing
Ioannis Avramopoulos and Jennifer Rexford, Princeton University

INVITED TALKS

Gold and Fool's Gold: Successes, Failures, and Futures in Computer Systems Research
Butler Lampson, Microsoft Research

People have been inventing new ideas in computer systems for nearly four decades, usually driven by Moore's Law. Many of them have been spectacularly successful: virtual memory, packet networks, objects, relational databases, and graphical user interfaces are a few examples. Other promising ideas have not worked out: capabilities, distributed computing, RISC, and persistent objects. And the fate of some is still in doubt: parallel computing, formal methods, and software reuse. The Web was not invented by computer systems researchers. In the light of all this experience, what will be exciting to work on in the next few years?

THE GURU IS IN

VoIP Security
Heison Chak, SOMA Networks Inc.

Heison is a system and network administrator who has become a VoIP fanatic. He will reveal his experience in VoIP systems security. VoIP solutions from open-source to commercial platforms will also be discussed.

5:30 p.m.–6:30 p.m. Friday
Plenary Session

Why Mr. Incredible and Buzz Lightyear Need Better Tools: Pixar and Software Development
Greg Brandeau, Vice President of Technology, Pixar Animation Studios

Movie-making and software are 100% interconnected at Pixar and other studios making computer-generated movies. This talk will discuss how we make movies at Pixar and how our movie-making process drives our software development efforts. We will discuss software requirements, development tools, and systems infrastructure. We will examine where we began and our current needs, as well as looking at the combinatorics that tell us what we will need in the future. We want to hear from you about what you think you can do, as we hope to engage the systems community to help make our vision a reality.




Greg Brandeau brings an extensive knowledge of engineering, technology, and management to his role as Vice President of Technology at Pixar Animation Studios. He and his group play a critical part in bridging all phases of the film production process at Pixar, providing the software tools, computers and audio-visual equipment used by the studio's artists to translate their creative visions to the big screen.
Greg first joined Pixar in 1996 as the studio's Director of Information Technology. After five successful years, during which he was promoted to vice president, Greg left the studio to broaden his expertise within other areas of technology. He returned to Pixar in 2004 as Vice President of Technology.
During Greg's time away from Pixar he served as Chief Information Officer for a Silicon Valley biotechnology startup, Perlegen Sciences. Earlier in his career he held a variety of senior level positions, including Director of Operations at NeXT and Director of Software Engineering at Mountain Network Solutions.
Greg earned Bachelor of Science and Masters of Science degrees in Electrical Engineering from the Massachusetts Institute of Technology in 1984. After serving in the U.S. Air Force, he continued his education at The Fuqua School of Business, Duke University, where he received a MBA in 1992.
Technical Sessions: Thursday, June 1 | Friday, June 2 | Saturday, June 3
Saturday, June 3
9:00 a.m.–10:30 a.m. Saturday
SYSTEMS PRACTICE & EXPERIENCE REFEREED PAPERS TRACK

Wide Area Distributed Systems

Service Placement in a Shared Wide-Area Platform
David Oppenheimer, University of California, San Diego; Brent Chun, Arched Rock Corporation; David Patterson, University of California, Berkeley; Alex C. Snoeren and Amin Vahdat, University of California, San Diego

Replay Debugging for Distributed Applications
Dennis Geels, Gautam Altekar, Scott Shenker, and Ion Stoica, University of California, Berkeley

Loose Synchronization for Large-Scale Networked Systems
Jeannie Albrecht, Christopher Tuttle, Alex C. Snoeren, and Amin Vahdat, University of California, San Diego

INVITED TALKS

Routing Without Tears, Bridging Without Danger
Radia Perlman, Sun Microsystems Laboratories

Why is route calculation done at both layers 2 and 3 of networking? Is one better? Do we need both? This talk explains the historical accident by which bridging was conceived and the properties that make it attractive, and dangerous, today. The talk discusses new work being done in IETF known as TRILL (TRansparent Interconnection of Lots of Links), which combines the advantages of bridges (layer 2 forwarding devices) and routers (layer 3 forwarding devices).

Dr. Radia Perlman is a Distinguished Engineer at Sun Microsystems Labs. Radia is the creator of the spanning tree algorithm used by the bridges (switches) of the Internet. She notes the irony that this work discussed in this talk might kill spanning tree bridging.

THE GURU IS IN

Privacy & Cryptography
Steven M. Bellovin, Columbia University

Privacy is a property of total system design. Steve will discuss things such as authentication, logging, databases, and cryptography as a way of achieving and preserving user privacy.

10:30 a.m.–11:00 a.m.   Break  
11:00 a.m.–12:30 p.m. Saturday
SYSTEMS PRACTICE & EXPERIENCE REFEREED PAPERS TRACK

Network and Operating System Support

System- and Application-level Support for Runtime Hardware Reconfiguration on SoC Platforms
Dimitris Syrivelis and Spyros Lalis, University of Thessaly, Hellas

Resilient Connections for SSH and TLS
Teemu Koponen, Helsinki Institute for Information Technology; Pasi Eronen, Nokia Research Center; Mikko Särelä, Helsinki University of Technology

Structured and Unstructured Overlays under the Microscope: A Measurement-based View of Two P2P Systems That People Use
Yi Qiao and Fabián E. Bustamante, Northwestern University

INVITED TALKS

An Introduction to Software Radio
Eric Blossom, Blossom Research; Coordinator and Maintainer of the GNU Radio Project

Software radio is a technique for building wireless communication systems (radios!) that gets the software as close to the antenna as possible. This talk describes software radio, how it differs from hardware radio, why it's possible now, and technical tricks involved in getting it to work, as well as applications of the technology.

THE GURU IS IN

Surviving Spam Mail
Gray Watson, MailNull.com

Gray is a software architect/engineer and web-tools developer who has spent years helping users avoid and control spam with his free MailNull.com system. He has extensive experience with UNIX mail servers, network applications, and appliances that classify, block, and combat spam. He will be available to discuss best practices in protecting your systems and accounts from spam using the latest protocols, software, and other anti-spam measures.

12:30 p.m.–2:00 p.m.   Lunch   (on your own)  
2:00 p.m.–3:30 p.m. Saturday
SYSTEMS PRACTICE & EXPERIENCE REFEREED PAPERS TRACK

Short Papers Session III

Reclaiming Network-wide Visibility Using Ubiquitous Endsystem Monitors
Evan Cooke, University of Michigan; Richard Mortier, Austin Donnelly, Paul Barham, and Rebecca Isaacs, Microsoft Research, Cambridge

Integrated Scientific Workflow Management for the Emulab Network Testbed
Eric Eide, Leigh Stoller, Tim Stack, Juliana Freire, and Jay Lepreau, University of Utah

How DNS Misnaming Distorts Internet Topology Mapping
Ming Zhang, Microsoft Research; Yaoping Ruan, IBM Research; Vivek Pai and Jennifer Rexford, Princeton University

Efficient Query Subscription Processing for Prospective Search Engines
Utku Irmak, Polytechnic University; Svilen Mihaylov, University of Pennsylvania; Torsten Suel, Polytechnic University; Samrat Ganguly and Rauf Izmailov, NEC Laboratories America

IP Only Server
Muli Ben-Yehuda, Oleg Goldshmidt, Elliot K. Kolodner, Zorik Machulsky, Vadim Makhervaks, Julian Satran, Marc Segal, Leah Shalev, and Ilan Shimony, IBM Haifa Research Laboratory

INVITED TALKS

Hackers and Founders
Paul Graham, Y Combinator

Hackers make better startup founders than MBAs. You don't need to know a lot about business to found a startup. Does that mean any good hacker can be a successful startup founder? Almost, but not quite. You need a few other things.

THE GURU IS IN

High Capacity and High Performance Storage on a Limited Budget
Jacob Farmer, Cambridge Computer Corporation

Jacob Farmer is the CTO for Cambridge Computer, an integrator and consulting group specializing in storage networking and data protection. Jacob has spent the past 15 years helping academic and research groups get the most of out of their storage dollar. In this session he will share examples and field questions about affordable solutions to massive storage problems.

3:30 p.m.–4:00 p.m.   Break  
4:00 p.m.–5:30 p.m. Saturday
Closing Session

Real Operating Systems for Real-time Motion Control
Trevor Blackwell, CTO, Anybots

Motion-control systems have traditionally used microcontrollers and minimal operating systems, but it is now practical to run real-time control systems on UNIX, write them in high-level languages such as Python, and use convenient UI environments such as Gnome. I'll demo some systems I've built, ranging from a self-balancing unicycle to a human-sized walking robot, and talk about how to implement them in ways similar to the e-commerce systems I used to build.

Technical Sessions: Thursday, June 1 | Friday, June 2 | Saturday, June 3
footer
?Need help? Use our Contacts page.

Last changed: 31 May 2006 ch