Listen to the opening remarks in MP3 format

Keynote Address
The Impact of Virtualization on Computing Systems
Mendel Rosenblum, Stanford University

Listen to the keynote in MP3 format

This talk describes how virtualization is changing the way computing is done in the industry today and how it is causing users to rethink how they view hardware, operating systems, and application programs. The talk will describe this new view of computing and the benefits driving users to adopt it. The roles of hardware and operating systems will be discussed, along with what changes will be needed to support this new computing model efficiently and simply.

Life Is Not a State-Machine: The Long Road from Research to Production
Werner Vogels, VP and CTO, Amazon.com

Listen in MP3 format

Traditionally a technology adaption cycle progresses at least 10–15 years before technologies become mature enough for wide-spread adaption. That time period is dramatically shortened as there is a need for technologies that can satisfy the unlimited appetite for ultra-scalable, highly-reliable, high-performance, and cost-effecient software architectures by the top Internet companies. In reality, however, it turns out to be very difficult to speed up the adoption process. In this presentation I will review some of the obstacles that are in the way of adoption of research results into production environments and will revist the principles of "worse is better" and Occam's razor in the context of technology transition.

Exploiting Online Games
Gary McGraw, Cigital

Listen in MP3 format

This talk (based on a book of the same title co-authored by Greg Hoglund) frankly describes controversial security issues surrounding MMORPGs such as World of Warcraft. This no-holds-barred approach is fully loaded with code examples, debuggers, bots, and hacks, of interest whether you are a gamer, a game developer, a software security person, or an interested bystander. I will cover:

• Why online games are a harbinger of software security issues to come
• How millions of gamers have created billion-dollar virtual economies
• How game companies invade your privacy
• Why some gamers cheat
• Techniques for breaking online game security
• How to build a bot to play a game for you
• Methods for total conversion and advanced mods

Second Life
Rob Lanphier and Mark Lentczner, Linden Lab

Listen in MP3 format: Talk | Q & A

Second Life is a 3D online world with a rapidly growing population from more than 100 countries around the globe. Residents create and build the world, which includes homes, vehicles, nightclubs, stores, landscapes, clothing, and games. In January 2007, Linden Lab (creators of Second Life) released the source code of the client ("viewer") application under the terms of the GNU General Public License.

Rob Lanphier, Linden Lab's Open Source Busybody, and Mark Lentczner, who directs a software engineering studio at Linden Lab, will talk about the release of the Second Life viewer source code: what that means, what it might mean, and what it doesn't mean. He'll provide a brief overview of the technology and history of Second Life, discuss the astronomical growth in use of Second Life, and explain what Linden Lab is doing to cope with the ever-increasing stress on the system. He'll discuss some key improvements Linden Lab is making in the protocols used by the product—utilizing a Web services model to increase scalability and to decouple versioning between clients and servers, as well as server-to-server communication.

Prior to starting with Linden Lab in September 2006, Rob Lanphier worked at a number of companies, including Microsoft, Asymetrix, Conjungi, and RealNetworks, as well as working as an independent consultant specializing in MediaWiki development. During his nine years at RealNetworks, Rob was a key contributor to two important multimedia standards (RTSP and SMIL) and was a leading force behind RealNetworks' open source initiative (Helix Community).

Mark Lentczner directs a software engineering studio at Linden Lab. His studio is primarily focused on the architectural extension of Second Life and the software infrastructure to support it. He has worked in Silicon Valley for over 20 years leading engineering teams on projects including virtual machines, software tools, cell phone browsers, and audio processing. Find out more here.

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 snacks. Check out the list of accepted posters and available abstracts.

Specializing General-Purpose Computing: A New Approach to Designing Clusters for High-Performance Technical Computing
Win Treese, SiCortex, Inc.

Listen in MP3 format: Talk | Q & A

High-performance technical computing stresses computer systems in many ways, from CPU performance to memory systems to inter-system communication. Over the past twelve years, clusters of commodity hardware running Linux have become the most common tool for high-performance computing. However, the dynamics of such applications are often very different from those of applications that drive the design of commodity computer systems, which means that commodity systems may be cheap for computing, but they are not efficient for many technical applications.

At the same time, it has become increasingly clear that the costs of software dominate the overall investment. Providing more efficient computing for the same software can therefore be extremely valuable in improving the delivered performance for technical applications. With Linux on clusters as the primary environment for such applications, we can consider how to design a system optimized for executing high-performance technical code on Linux. SiCortex has recently introduced one example of such a system. The SC5832 and SC648 are computer systems designed from the silicon up to run Linux cluster applications. The systems feature a large number of processor cores, integrated high-speed interconnect fabric, a small footprint, and low power usage. In essence, the hardware design has taken the important components for a cluster node and shrunk them onto a single chip (plus DRAM); the software provides a coherent, turnkey Linux cluster computing environment.

We will examine the technology trends that converge to make this design approach both feasible and attractive, including the role of lowering power consumption as a means of gaining performance. In addition, we will describe the design of individual processor nodes, the interconnect fabric, and the challenges of assembling a Linux cluster distribution.

Live Malware Attack!
Paul Ducklin, Sophos

Listen in MP3 format

If you are interested in the source code of tools used during the demo, please email conference@usenix.org for more information.

Pretty much everyone knows that there is a lot of hostile code out there. But how does a malware attack unfold? How do exploits work in real life? How easily might you be phished? More important, how do you deliberately go about watching malware at work (in both senses of "at work"), especially if it is malware that relies on the Internet to function?

This talk will feature a live—but entirely self-contained, and therefore safe!—demonstration of a modern malware attack in action. Gain insight into how the bad guys think and operate, and you learn how better to defend yourself against them.

The talk will also examine some of the tricks and techniques that can be used in a malware research lab to get even an apparently complex and heavily obfuscated piece of malware to reveal its secrets in safety.

Paul Ducklin works for Sophos in Australia, where he does assorted security stuff. He is Head of Technology, Asia Pacific. If you find out what this means, please attend his talk so that you can tell him.

LiveJournal's Backend Technologies
Brad Fitzpatrick, LiveJournal

Listen in MP3 format: Talk | Q & A

Hear the history and lessons learned while scaling a community site (LiveJournal.com) from a single server with a dozen friends to hundreds of machines and 10M+ users: what's worked, what hasn't, and all the things we've had to build ourselves that are now in common use thoughout the Web 2.0 world, including memcached, MogileFS, Perlbal, and our job dispatch systems.

Brad Fitzpatrick created LiveJournal in 1999 and grew the company throughout and after college, later selling it to SixApart, creators of TypePad, MovableType, and Vox. The open souce infrastructure software created to keep LiveJournal alive throughout the years is now popular within the Web 2.0 world. Brad is also responsible for creating OpenID, originally designed for interop among SixApart Web sites.

MapReduce and Other Building Blocks for Large-Scale Distributed Systems at Google
Jeffrey Dean, Google

Listen in MP3 format: Talk | Q & A

MapReduce is a programming model and an associated implementation for processing and generating large data sets. Users specify a Map function which processes a key/value pair to generate a set of intermediate key/value pairs, and a Reduce function which merges all intermediate values associated with the same intermediate key. Programs written in this functional style are automatically parallelized and executed on a large cluster of commodity machines. The MapReduce run-time system takes care of the details of partitioning the input data, scheduling the program's execution across a set of machines, handling machine failures, and managing the required inter-machine communication. This allows programmers without any experience with parallel and distributed systems to easily utilize the resources of a large distributed system. Thousands of MapReduce programs have been implemented, and several thousand MapReduce jobs are executed on Google's clusters every day. In this talk I'll describe the design and implementation of MapReduce and other building blocks for large-scale distributed systems at Google.

Jeff is a Google Fellow in Google's Systems Infrastructure Group. While at Google, he has worked on Google's crawling, indexing, query serving, and advertising systems and has built various pieces of Google's distributed computing infrastructure. Prior to joining Google, he was at DEC/Compaq's Western Research Laboratory, where he worked on profiling tools, microprocessor architecture, and information retrieval. Prior to graduate school, he worked at the World Health Organization's Global Programme on AIDS.

Perfect Data in an Imperfect World
Daniel V. Klein, Consultant

Listen in MP3 format: Talk | Q & A

It is no secret that we are at the dawn of the digital age—our parents (and, for some of us, even our grandparents) have computers, digital cameras, MP3 players, etc. We each have more computing power in our cell phones than the mainframes of 35 years ago had, and everywhere we find data acquisition and tracking systems. Privacy has never been more zealously guarded nor more freely abandoned, and with the proliferation of digital data collection and dissemination have come new worries.

What is being recorded, why, and by whom? With literally billions of computers around us, how can we keep our data (and ourselves) safe? How can we prevent misappropriation or misuse of information about ourselves? How can we ever expunge flawed records, urban legends, or embarrassing facts? We have become the elephant who never forgets, but what are we remembering?

This talk will take a look at what our world is becoming, and perhaps suggest what we can do to make it a little less imperfect.

Dan Klein began his life of crime in 2nd grade, when he was caught with a pack of firecrackers. Since then his brushes with authority have been sporadic but relentless, but have not managed to deny him a security clearance, a job, or his well-deserved reputation as an off-the-wall maverick. His computer experience has included simulation and process control, the internals of almost every UNIX kernel released in the past 28 years, and graphical user interface management systems.

Human Computation
Luis von Ahn, Carnegie Mellon University

Listen in MP3 format: Talk | Q & A

Tasks like image recognition are trivial for humans, but continue to challenge even the most sophisticated computer programs. This talk introduces a paradigm for utilizing human processing power to solve problems that computers cannot yet solve. Traditional approaches to solving such problems focus on improving software. I advocate a novel approach: constructively channel human brainpower using computer games. For example, the ESP Game, described in this talk, is an enjoyable online game—many people play over 40 hours a week—and when people play, they help label images on the Web with descriptive keywords. These keywords can be used to significantly improve the accuracy of image search. People play the game not because they want to help, but because they enjoy it. The ESP Game has been licensed by a major Internet company and will soon become the basis of their image search engine.

In addition, I describe my work on CAPTCHAs, automated tests that humans can pass but computer programs cannot. CAPTCHAs take advantage of human processing power in order to differentiate humans from computers, an ability that has important applications in practice.

The results of this work are currently in use by hundreds of Web sites and companies around the world, and over 300,000 people have played some of the games presented here. Practical applications include improvements in areas such as computer vision, image search, adult-content filtering, spam prevention, common-sense reasoning, accessibility, and security in general.

Luis von Ahn is an assistant professor in the Computer Science Department at Carnegie Mellon University, where he received his Ph.D. in 2005. Luis obtained a B.S. in mathematics from Duke University in 2000. He is the recipient of a MacArthur Fellowship and was named one of Popular Science magazine's "Brilliant 10" scientists of 2006. His research interests include encouraging people to do work for free, as well as catching and thwarting cheaters in online environments.

Warehouse-scale Computers
Luiz André Barroso, Google Inc.

Listen in MP3 format: Talk | Q & A

The computing systems that are powering many of today's large-scale Internet services look less like refrigerators and more like warehouses. Designing efficient warehouse-scale computers requires many of the traditional tools and methods developed by computer architects, and some new tricks as well. In this talk I'll describe some of the defining characteristics of these systems, with a focus on failure handling and power management.

Luiz André Barroso is a Distinguished Engineer at Google, where he has worked across several engineering areas, ranging from applications and software infrastructure to hardware design. Prior to working at Google, he was a member of the Research Staff at Compaq and Digital Equipment Corporation, where his group did some of the pioneering work on computer architectures for commercial workloads. That work included the design of Piranha, a system based on an aggressive chip-multiprocessing, which helped inspire many of the multi-core CPUs that are now in the mainstream.

Luiz has a Ph.D. degree in computer engineering from the University of Southern California and B.S. and M.S. degrees in electrical engineering from the Pontifícia Universidade Católica, Rio de Janeiro.

Crossing the Digital Divide: The Latest Efforts from One Laptop per Child
Mary Lou Jepsen, One Laptop per Child

Listen in MP3 format: Talk | Q & A

A lot of people talk about the digital divide—I wanted to do something about it.

This effort emerged as a way to capture the endless momentum of Moore's Law and create a laptop for those far on the other side of the digital divide—the poor children of the world and their families. In fact, the vast majority of the world lives without so many of the things we consider essential, not least of which is access to education and information. This year, we intend to launch with millions of laptops simultaneously in Rwanda, Pakistan, Brazil, Argentina, Uruguay, Libya, Nigeria, and Thailand. The children themselves will own these laptops, which will be distributed to them by the Ministries of Education. They should last for five years and are cheaper than five years' worth of textbooks in the average developing country.

We have not created a cost-reduced version of today's laptop; we have created an entirely new approach to the idea of a laptop, which I will discuss in some detail. It has three significant features you would like in your own laptop: (1) it is so low-power that it can make its own electricity, which is critical in a world where 50% of the kids have little or no access to electricity at home; (2) through mesh networking, it is its own hot spot; (3) it has a sunlight readable screen, which is key when so often in the developing world school is held under a tree. As we start mass production, many details of deployment, teacher preparation, content, in-country maintenance, and in-country networking are being finalized. I will discuss our latest efforts and plans in these areas.

Mary Lou Jepsen is co-founder and chief technology officer of One Laptop per Child, where she is responsible for overall hardware management. Previously she was a group executive at Intel and a co-founder and CTO of the MicroDisplay Corporation. She is widely regarded as a display technology pioneer. She is also a professor at the MIT Media Lab.