TECHNICAL SESSIONS
Now available! Check out the videos, MP3s, and slides of the presentations. (Take them with you—they're iPhone compatible!)
Conference papers are available to conference registrants immediately and to everyone beginning Wednesday, November 4. Everyone can view the proceedings front matter immediately.
Proceedings Front Matter:
Title Page |
Conference Organizers and External Reviewers |
Table of Contents |
Message from the Program Chair
Complete Proceedings (PDF)
Tech Sessions:
Wednesday, November 4 |
Thursday, November 5 |
Friday, November 6 |
Invited Talk Speakers
Wednesday, November 4
|
8:45 a.m.–10:30 a.m.
|
Wednesday
|
Harborside Ballroom C
Opening Remarks, Best Paper Awards, and 2009 SAGE Outstanding Achievement Award
Adam Moskowitz, LISA '09 Program Chair
View the video
Listen in MP3 format
View the presentation slides
Keynote Address
Ahead in the Cloud: The Power of Infrastructure as a Service
Werner Vogels, CTO, Amazon.com
Building a reliable, flexible infrastructure that can scale up or down at a moment's notice can be a complicated, expensive challenge, but it's essential for businesses today, whether they are enterprises trying to cut costs, young businesses saturated with customer demand, or start-ups about to launch. This talk will review lessons learned from building Amazon.com, one of the world's largest distributed systems. It will focus on state management, a key factor for scalability, reliability, performance, and cost-effectiveness.
|
10:30 a.m.–11:00 a.m. Break with Drinks and Snacks |
Harborside Ballroom Foyer |
|
11:00 a.m.–12:30 p.m. |
Wednesday |
REFEREED PAPERS
Harborside Ballroom A & B
The Human Side of Sysadmin
Session Chair: John Sellens, SYONEX
Pushing Boulders Uphill: The Difficulty of Network Intrusion Recovery
Michael E. Locasto, George Mason University; Matthew Burnside, Columbia University; Darrell Bethea, University of North Carolina at Chapel Hill
Paper in PDF | Slides
Two-Person Control Administration: Preventing Administation Faults through Duplication
Shaya Potter, Steven M. Bellovin, and Jason Nieh, Columbia University
Paper in PDF
View the video
Listen in MP3 format
The Water Fountain vs. the Fire Hose: An Examination and Comparison of Two Large Enterprise Mail Service Migrations
Craig Stacey, Max Trefonides, Tim Kendall, and Brian Finley, Argonne National Laboratory
Paper in PDF | Slides
View the video
Listen in MP3 format
|
INVITED TALKS
Harborside Ballroom C
Session Chair: Paul Armstrong
How to Build a PB Sized Disk Storage System
Raymond L. Paden, HPC Technical Architect, IBM Deep Computing
View the video
Listen in MP3 format
View the presentation slides
At one time, petabyte-sized disk storage systems seemed unthinkable, but today they are increasingly common. Vendors are responding by making larger, less expensive disks and controllers that manage over a PB of data. Yet center managers and application developers do not truly appreciate the complexity of these large storage systems. Too often they approach them as being peripheral rather than integral to the overall processing system. They give detailed attention to other things, but simply ask for a storage system like the ones they had before, only bigger. This often leads to an overly costly storage system that is a performance bottleneck. This talk will help you avoid the pitfalls and develop an efficient PB-sized system.
|
THE GURU IS IN
Essex A, B, & C
11:00 a.m.–11:45 a.m.
Zenoss
Chet Luther, Zenoss, Inc.
Chet Luther has been in the system, network, and security administration business for the past ten years. He has worked with everything from home-grown monitoring tools for small data centers to NetSaint/Nagios deployments on nationwide networks. Throughout this odyssey he always struggled with the difficulty of using open tools to monitor heterogeneous environments across all layers. These struggles led him to try to replace an existing Nagios installation with Zenoss Core, which in turn led him to heavy involvement in the community and, eventually, to his current role as a Client Services Engineer at Zenoss, Inc. These days he works with the community and with customers managing from 20 to 20,000 devices to find better solutions to their network management woes.
11:45 a.m.–12:30 p.m.
Nagios and Unnoc
Kyle Martin and Adam Augustine, The Church of Jesus Christ of Latter-day Saints
Kyle Martin is a Monitoring Engineer for The Church of Jesus Christ of Latter-day Saints. He specializes in providing solutions for the monitoring of applications, servers, and appliances using SOAP, Perl, and bash. Almost all monitoring solutions currently used at the LDS Church are open source.
Adam Augustine is the Monitoring Team Manager for The Church of Jesus Christ of Latter-day Saints. Over his 16-year career he has specialized in building low-cost and scalable low-level infrastructure and its supporting systems. He has been working on the Monitoring team at the Church for five years.
|
|
12:30 p.m.–2:00 p.m. Lunch (on your own) |
|
2:00 p.m.–3:30 p.m. |
Wednesday |
INVITED TALKS I
Harborside Ballroom A & B
Session Chair: Narayan Desai, Argonne National Laboratory
Eucalyptus: An Open Source Infrastructure for Cloud Computing
Rich Wolski, Chief Technology Officer and co-founder of Eucalyptus Systems Inc. and Professor of Computer Science at the University of California, Santa Barbara
View the video
Listen in MP3 format
View the presentation slides
We will present Eucalyptus (Elastic Utility Computing Architecture for
Linking Your Programs to Useful Systems), an open source software
infrastructure that implements IaaS-style cloud computing. The goal of
Eucalyptus is to allow sites with existing clusters and server infrastructure
to host a cloud that is interface-compatible with Amazon's AWS and (soon) the
Sun Cloud open API. In addition, through its interfaces Eucalyptus is able to
host cloud platform services such as AppScale (an open source implementation of
Google's AppEngine) and Hadoop, making it possible to "mix and match"
different service paradigms and configurations within the cloud. Finally,
Eucalyptus can leverage a heterogeneous collection of
virtualization technologies within a single cloud, to
incorporate resources that have already been virtualized without modifying
their configuration.
This talk will focus on specific features of the system that are designed to
enable rapid development, prototyping, and deployment of local computing
clouds, particularly for debugging and/or application development. It
will also cover experiences of hosting the Eucalyptus Public Cloud (EPC) as
a free public cloud platform for experimental use, as well as the ability to use the
EPC in conjunction with commercial Web development services,
such as Rightscale, that target AWS. Finally, it will describe what it was like to build and
support such an open source cloud infrastructure and will point to potential
directions that are likely to enable further innovation.
|
INVITED TALKS II
Harborside Ballroom C
Session Chair: Mario Obejas, Raytheon
The Advanced Persistent Threat
Michael K. Daly, Director of Enterprise Security Services, Raytheon Company
View the video
Listen in MP3 format
View the presentation slides
Critical infrastructures and the governments, corporations, and individuals supporting them are under attack by increasingly sophisticated cyber threats from hostile entities. Their goal is to gain access to intellectual property, personally identifiable information, financial data, and targeted strategic information. This is not simple fraud or hacking. It is intellectual property theft and infrastructure corruption on a grand scale. This talk will discuss the ways operators of national critical infrastructure are working to combat these threats and the collaborative partnerships that have been formed to strengthen our collective defenses.
|
THE GURU IS IN
Essex A, B, & C
2:00 p.m.–2:45 p.m.
Web Site Scaling
Matthew Barr, Interactive One
Matthew Barr is a Senior Systems Engineer at Interactive One (BlackPlanet.com, The Urban Daily, Hello Beautiful). While there, he has worked on projects ranging from a complete datacenter rebuild to architecting and implementing configuration management and other infrastructure services. He is currently focused on tuning the various systems, from Web servers to MySql databases.
2:45 p.m.–3:30 p.m.
Data Centers
Doug Hughes, D.E. Shaw Research
Doug Hughes is the technical lead for systems and operations at a privately held chemistry research company in Manhattan. His responsibilities include datacenter architecture and maintenance, storage, networking, and everything related to commodity computing and HPC. His team is also responsible for integrating an in-house–built, custom supercomputer for computational chemistry. Prior to his current employment Doug worked at Global Crossing as a system administrator supporting the global IP network, and prior to that as the senior system administrator for the College of Engineering at Auburn University. Doug has a B.E. in Computer Engineering from Pennsylvania State University.
|
|
3:30 p.m.–4:00 p.m. Break with Drinks and Snacks Vendor Exhibition, Grand Ballroom Salons VI–X |
|
4:00 p.m.–5:30 p.m. |
Wednesday |
Harborside Ballroom C
Plenary Session
Session Chair: William LeFebvre, Digital Valence, LLC
Google Wave
Daniel Berlin and Joe Gregorio,
Google, Inc.
View the video
Listen in MP3 format
View the presentation slides
Google Wave, a new tool for communication and collaboration on the
Web, leverages an innovative real-time multi-user document known
as a Wave. Each Wave is a hosted document that can be used for various
idiomatic workflows including, but not limited to, instant-messaging-style conversation, document collaboration, photo sharing, event
organizing, live blogging, meeting agendas, flame wars, or a
combination of all of the above.
In the simplest form, each Wave contains a hierarchy of
named XML documents. Each of these documents supports concurrent
modifications and low-latency updates between their participants. The
Google Wave Federation Protocol is an open extension to XMPP that
defines how these hosted documents are shared
between organizations and their users (typically bounded via
domain name, in the same style as email or an instant-messaging
service). This XMPP extension is an almost completely new type of protocol,
one that is far from the delayed nature of SMTP and requires a
greater level of reliability than traditional instant messaging. More
broadly, this new protocol poses various challenges for development
and deployment, some of which may even be seen as hurdles to
mainstream adoption.
This talk will present the Google Wave Federation Protocol and discuss its relationship to existing systems and networks.
|
|
5:30 p.m.–6:30 p.m. |
Wednesday |
Exhibit Hall Happy Hour
Vendor Exhibition, Grand Ballroom Salons VI–X
|
6:30 p.m.–7:30 p.m. |
Wednesday |
Grand Ballroom Salon Foyer (outside VI–X)
Poster Session
Session Chair: Gautam Singaraju, The University of North Carolina at Charlotte
Don't miss the cool new ideas and the latest preliminary research on display at the Poster Sessions. Take advantage of a perfect opportunity to mingle with colleagues who may be interested in the same area.
Check out the list of accepted posters.
|
|
Tech Sessions:
Wednesday, November 4 |
Thursday, November 5 |
Friday, November 6 |
Invited Talk Speakers
|
|
Thursday, November 5 |
9:00 a.m.–10:30 a.m.
|
Thursday
|
Harborside Ballroom C
Plenary Session
Session Chair: Andrew Hume, AT&T Labs—Research
Cosmic Computing: Supporting the Science of the Planck Space Based Telescope
Shane Canon,
Group Leader, Data System Group in NERSC, Lawrence Berkeley National Laboratory
View the video
Listen in MP3 format
View the presentation slides
The scientific community is creating data at an ever-increasing rate. Large-scale experimental devices such as high-energy collider facilities and advanced telescopes generate petabytes of data a year. These immense data streams stretch the limits of the storage systems and of their administrators. The Planck project, a space-based telescope designed to study the Cosmic Microwave Background, is a case in point. Launched in May 2009, the Planck satellite will generate a data stream requiring a network of storage and computational resources to store and analyze the data. This talk will present an overview of the Planck project, including the motivation and mission, the collaboration, and the terrestrial resources supporting it. It will describe the data flow and network of computer resources in detail and will discuss how the various systems are managed. Finally, it will highlight some of the present and future challenges in managing a large-scale data system.
|
10:30 a.m.–11:00 a.m. Break with Drinks and Snacks Vendor Exhibition, Grand Ballroom Salons VI–X |
|
11:00 a.m.–12:30 p.m. |
Thursday |
INVITED TALKS I
Harborside Ballroom A & B
Session Chair: Doug Hughes, D.E. Shaw Research, LLC
Storage and Analysis Infrastructure for Anton
Mark Moraes, Head of Anton Software Development and Systems Software, D.E. Shaw Research
Anton is a recently completed special-purpose supercomputer that
accelerates molecular dynamics simulations of biological systems by
orders of magnitude compared with the previous state of the art. As a
result of this dramatic increase in simulation speed, each Anton machine
is capable of producing terabytes of results ("trajectories") per day.
Such large trajectories pose significant networking, storage, and
computational challenges. This talk will describe how we tackled these
challenges when creating scalable storage and analysis infrastructure
for Anton.
|
INVITED TALKS II
Harborside Ballroom C
Session Chair: Gautam Singaraju, The University of North Carolina at Charlotte
Is ITIL(r) All Theory and No Practice?
Carolyn M. Hennings, Windward IT Solutions
View the presentation slides
IT organizations exist to provide computing resources and services needed by their customers. Customers range from businesses to academic organizations and government agencies, each having their own unique requirements. The technical practices discussed at this conference would not be needed were it not for customer needs and requirements. These customers pay our salaries. They pay for the hardware, the software, the datacenter space, and everything else we need to do our jobs. In return, they expect IT services. IT Service Management—specifically, ITIL(r)—is simply a way of looking at how we provide those services and determining whether we are doing it in the most effective and efficient ways.
This talk provides an overview of the IT Service Management lifecycle described in version 3 of the IT Infrastructure Library. Through references to papers and invited talks from previous LISA conferences, this talk offers concrete examples of the work we do in the context of the IT service lifecycle stages: Service Strategy, Service Design, Service Transition, Service Operation, and Continual Service Improvement.
|
THE GURU IS IN
Essex A, B, & C
VMware
John Arrasjid and Rupen Seth, VMware
John Arrasjid (VCDX, VCP, ITIL Foundations) is a Principal Architect at VMware, where he has developed tools, training material, design strategies, and content for the VCDX certification program. John is co-author of the Short Topics booklet Deploying the VMware Infrastructure, which focuses on the VI 3.x environment. John is currently developing a new book, to be titled VMware Architecture Design Patterns: Blueprints for Building the Cloud, scheduled for release in 2009. John has focused his virtualization work on disaster recovery, performance, and security. John has run tutorials on VMware technology at USENIX Annual Technical and LISA conferences since 2005, in the areas of VMware Infrastructure 2/3, performance, and advanced topics. Prior to working at VMware, John held IT consulting, architecture, developer, and management roles at WebNexus, Roxio, 3Dfx, AT&T, and Chronologic Simulation. John holds his Bachelor of Science in Computer Science from the State University of New York (SUNY) at Buffalo.
Rupen Sheth (VCP, ITIL Foundations) is a Worldwide Consulting Services Architect at VMware. Rupen is currently responsible for the technical accuracy, development, and release of numerous materials to enable VMware field consultants and partners to deliver quality services using VMware solutions. Rupen is currently involved in developing service delivery content for vSphere and vCenter management solutions. Rupen is co-author of the book The Path to VMware vSphere—Unleashed and has presented at numerous VMware shows and events. Before joining VMware, Rupen fulfilled numerous consulting, architect, and engineering roles at BearingPoint, Citigroup, American Airlines, and GE Medical Systems. Rupen has a B.S. in Electrical Engineering and an M.S. in Computer Science from the University of Wisconsin.
|
|
12:30 p.m.–2:00 p.m. Lunch (on your own) |
|
2:00 p.m.–3:30 p.m. |
Thursday |
REFEREED PAPERS
Harborside Ballroom A & B
Networks, Networks, Networks
Session Chair: Mario Obejas, Raytheon
Awarded Best Paper!
Crossbow Virtual Wire: Network in a Box
Sunay Tripathi, Nicolas Droux, Kais Belgaied, and Shrikrishna Khare, Sun Microsystems, Inc.
Paper in PDF | Slides
View the video
Listen in MP3 format
EVA: A Framework for Network Analysis and Risk Assessment
Melissa Danforth, California State University, Bakersfield
Paper in PDF | Slides
View the video
Listen in MP3 format
An Analysis of Network Configuration Artifacts
David Plonka and Andres Jaan Tack, University of Wisconsin—Madison
Paper in PDF | Slides
View the video
Listen in MP3 format
|
INVITED TALKS
Harborside Ballroom C
Session Chair: Chris McEniry, Sony Computer Entertainment America
Searching for Truth, or at Least Data: How to Be an Empiricist Skeptic
Elizabeth D. Zwicky
View the video
Listen in MP3 format
View the presentation slides
What do software testing, security, and successful project planning have in common? They all require the same outlook. Call it "skeptical empiricist," call it "man from Missouri" (the Show Me state), call it data-driven, but whatever you call it, it involves looking beyond claims and guesses and trying to figure out what the facts are. This talk will provide you with tools and advice on how to find data and make sense of it, plus of course inspiring tales of triumph, disaster, and comedy revolving around the search for facts.
|
THE GURU IS IN
Essex A, B, & C
Amanda Open Source Backup
Nick Brockner, Hamilton College; Dustin Mitchell, Zmanda, Inc.
Nick Brockner is an LPIC-1 certified system administrator who holds a B.S. in Computer and Information Science and is working on an M.S. in Computer Science. He has extensive hands-on, self-directed and self-taught experience in backups in pure Windows environments, mixed *nix/Wintel, and pure Linux environments. Amanda Open Source Backup has saved both him and his employer time, money, and headaches. The transparent nature of Amanda allows much more control and insight into what is really going on with your backups.
|
|
3:30 p.m.–4:00 p.m. Break with Drinks and Snacks |
Harborside Ballroom Foyer |
|
4:00 p.m.–5:30 p.m. |
Thursday |
REFEREED PAPERS
Harborside Ballroom A & B
Security, Security, Security
Session Chair: David Plonka, University of Wisconsin
Secure Passwords Through Enhanced Hashing
Benjamin Strahs, Chuan Yue, and Haining Wang, The College of William and Mary
Paper in PDF | Slides
View the video
Listen in MP3 format
SEEdit: SELinux Security Policy Configuration System with Higher Level Language
Yuichi Nakamura and Yoshiki Sameshima, Hitachi Software Engineering Co., Ltd.; Toshihiro Tabata, Okayama University
Paper in PDF | Slides
View the video
Listen in MP3 format
An SSH-based Toolkit for User-based Network Services
Joyita Sikder, University of Illinois at Chicago; Manigandan Radhakrishnan, VMware; Jon A. Solworth, University of Illinois at Chicago
Paper in PDF
View the video
Listen in MP3 format
|
INVITED TALKS
Harborside Ballroom C
Session Chair: Ian Dotson, University of Washington
Delay/Disruption Tolerant Networking
Keith Scott, MITRE Corporation
View the video
Listen in MP3 format
View the presentation slides
This talk will present an overview of Delay/Disruption Tolerant Networking (DTN), including its heritage in the InterPlaNetary Internet, its current capabilities, ongoing work on DTN by the Internet Research Task Force (IRTF) and the Consultative Committee for Space Data Systems (CCSDS), and DTN's applicability to NASA and international space missions. The discussion will focus on the naming and routing aspects of the Bundle Protocol (RFC5050), including how it can be used to support features such as delayed binding and content-based networking.
|
THE GURU IS IN
Essex A, B, & C
4:00 p.m.–4:45 p.m.
Building and Maintaining Secure Systems
Kevin A. Nassery,
Consciere LLC
Kevin Nassery is a hands-on technical architect who has been a UNIX system, network, and security engineer and consultant for over a decade. After serving for more than four years as principal infrastructure architect for a major online presence, he recently returned to his passion of security consulting. He is an RHCE, a CISSP, and a graduate student at Depaul University, where he studies computer, information, and network security. He is currently a senior security consultant with Consciere LLC.
4:45 p.m.–5:30 p.m.
Network Management
D. Brent Chapman,
Netomata, Inc.
Brent Chapman is the founder and manager of the Network-Automation mailing list and the creator of the Netomata Config Generator (NCG) open source software, which automates the generation of coordinated, complete, and correct config files for network devices and services. He has over 20 years of information technology management experience, in organizations large and small, much of it focused on network management and automation. He is the coauthor of the highly regarded O'Reilly & Associates book Building Internet Firewalls and the creator of the Majordomo mailing list management package. In 2004 Brent was honored with the SAGE Outstanding Achievement Award "for outstanding sustained contributions to the community of system administrators."
|
|
6:00 p.m.–7:00 p.m. |
Thursday |
Grand Ballroom Salon Foyer (outside VI–X)
Poster Session
Session Chair: Gautam Singaraju, The University of North Carolina at Charlotte
Don't miss the cool new ideas and the latest preliminary research on display at the Poster Sessions. Take advantage of a perfect opportunity to mingle with colleagues who may be interested in the same area.
Check out the list of accepted posters.
|
7:00 p.m.–8:30 p.m. |
Thursday |
Conference Reception: Halloween II
Grand Ballroom Salons VI–X |
|
Tech Sessions:
Wednesday, November 4 |
Thursday, November 5 |
Friday, November 6 |
Invited Talk Speakers
|
|
Friday, November 6 |
9:00 a.m.–10:30 a.m. |
Friday |
Harborside Ballroom C
Plenary Session
Session Chair: William LeFebvre, Digital Valence, LLC
Towards Zero-Emission Datacenters Through Direct Reuse of Waste Heat
Bruno Michel,
IBM Zurich Research Laboratory
View the video
Listen in MP3 format
View the presentation slides
High-performance liquid cooling allows datacenters to operate with coolant temperatures above the free cooling limit in all climates, eliminating the need for chillers and allowing the thermal energy to be reused in cold climates. We have demonstrated removal of 85% of the heat load from high-performance compute nodes at a temperature of 60 degrees Celsius and compared their energy and emission balance with a classical air-cooled datacenter, a datacenter with free cooling in a cold climate zone, and a datacenter with chiller mediated energy reuse. This talk will explain how our method reduces energy consumption by almost a factor of two compared to a current datacenter and reduces energy cost and carbon footprint by an even larger factor.
|
|
10:30 a.m.–11:00 a.m. Break with Drinks and Snacks |
Harborside Ballroom Foyer |
|
11:00 a.m.–12:30 p.m. |
Friday |
REFEREED PAPERS
Harborside Ballroom A & B
On the Fringe
Session Chair: Mark D. Roth, Google, Inc.
Awarded Best Student Paper!
Federated Access Control and Workflow Enforcement in Systems Configuration
Bart Vanbrabant, Thomas Delaet, and Wouter Joosen, K.U. Leuven, Belgium
Paper in PDF | Slides
View the video
Listen in MP3 format
CIMDIFF: Advanced Difference Tracking Tool for CIM Compliant Devices
Ramani Routray, IBM Almaden Research Center; Shripad Nadgowda, IBM India Systems and Technology Lab
Paper in PDF | Slides
View the video
Listen in MP3 format
Transparent Mobile Storage Protection in Trusted Virtual Domains
Luigi Catuogno and Hans Löhr, Ruhr-University Bochum, Germany; Mark Manulis, Technische Universität Darmstadt, Germany; Ahmad-Reza Sadeghi and Marcel Winandy, Ruhr-University Bochum, Germany
Paper in PDF | Slides
View the video
Listen in MP3 format
|
INVITED TALKS
Harborside Ballroom C
Session Chair: Doug Hughes, D.E. Shaw Research, LLC
Visualizing DTrace: Sun Storage 7000 Analytics
Bryan Cantrill, Sun Microsystems
View the video
Listen in MP3 format
System administrators have long been unable to meaningfully visualize
the run-time performance of their systems. This is not (merely) an
interface problem—the data available to visualize has not
historically
extended beyond basic counters and the ability to visualize the
performance of the system has therefore not extended much beyond simple
graphing of these scalar values. However, with the introduction of
DTrace
and its ability to perform ad hoc instrumentation of production systems,
the data available to visualize has become essentially limitless, with
the
possibilities only constrained by our ability to build facilities that can
ask the right questions and cogently display the answers. This talk
describes one such DTrace-based system visualization facility: the
appliance analytics found in the new Sun Storage 7000 series. This talk will demonstrate the facility, explain its innards, and provide examples of the ways visualization has allowed new answers to old questions and has itself posed new questions about our systems.
|
THE GURU IS IN
Essex A, B, & C
Splunk
Christina Noren and Patrick Ogdin, Splunk
Christina is the VP Product Management at Splunk. With a unique approach to blending support and product management into a single team, she's built systems and processes to capture every Splunk user interaction and synthesize them into interesting problems for engineering to solve. Prior to Splunk, Christina was the group manager responsible for MSN's monitoring infrastructure. She was a key early employee at SenSage, Portal Software, and Sonic Solutions and was co-founder and CEO for Artloop. Christina holds a B.A. in International Finance and Economics and a B.F.A. from the Dominican University of California.
Patrick is a Solutions Architect for Splunk working with customers in the Mid-Atlantic and Southeast. Prior to Splunk, Patrick was a Technical Specialist at Sun Microsystems for X64 servers and Solaris on X64 as well as a Systems Engineer in the New York Metro area. Patrick holds a B.S. in Systems Engineering from George Mason University.
|
|
12:30 p.m.–2:00 p.m. Lunch (on your own) |
|
2:00 p.m.–3:30 p.m. |
Friday |
INVITED TALKS I
Harborside Ballroom A & B
Session Chair: Paul Armstrong
E Unum Pluribus: Google Network Filtering Management
Paul (Tony) Watson and Peter Moody,
Google
View the video
Listen in MP3 format
View the presentation slides
Network filtering can be a very difficult challenge in large, complex, and sprawling networks. Through the use of internally developed software, Google has automated and simplified many of these difficult tasks and provided the capability to easily audit and validate its filters. We will discuss Google's efforts in this area and release some of these tools to the community.
|
INVITED TALKS II
Harborside Ballroom C
Session Chair: Paul Anderson, University of Edinburgh
Above the Clouds: A Berkeley View of Cloud Computing
Armando Fox,
Reliable Adaptive Distributed Systems Lab, University of California, Berkeley
View the video
Listen in MP3 format
View the presentation slides
Not for several years has a new concept generated so much noise in the
blogosphere as "cloud computing" has in the past several months. Is this
just hype, or is it actually something new? Is this another flash in
the pan, or is it a long-term paradigm shift (ugh)? This talk will
propose some informed answers to those questions (specifically, "yes, it is new"
and "yes, it is long-term"), will identify challenges as well as opportunities for the
continued growth of cloud computing, and will relate these to our own
experience as cloud computing users in our research and teaching
activities.
|
THE GURU IS IN
Essex A, B, & C
Interviewing and Job Hunting Skills
Thomas A. Limoncelli, Google NYC, and Adam Moskowitz, Menlo Computing
Thomas A. Limoncelli, author of The Art of Time Management for System Administrators from O'Reilly and co-author of the new 2nd edition of The Practice of System and Network Administration from Addison-Wesley, is a system administrator at Google in NYC. In 2005 he received the SAGE Outstanding Achievement Award, and he is a frequent presenter at LISA conferences. A sysadmin and network wonk since 1987, he previously worked at Cibernet, Dean for America, Lumeta, Bell Labs/Lucent, and Drew University. He blogs at http://EverythingSysadmin.com.
Adam Moskowitz has at various times been a programmer, a system
administrator, a manager of sysadmins, and a technical trainer. He has
been a Guru and speaker at past LISA conferences, has taught several
LISA tutorials about interviewing, has run the LISA Advanced Topics
Workshop for the past 12 years, and is the Program Chair of this year's
conference. Adam doesn't blog, but he does maintain a Web page of cute
pictures of his dog Ancho.
|
|
3:30 p.m.–4:00 p.m. Break with Drinks |
Harborside Ballroom Foyer |
|
4:00 p.m.–5:30 p.m. |
Friday |
Harborside Ballroom C
Plenary Closing Session
Session Chair: Adam Moskowitz
Frank Lloyd Wright Was Right!
Daniel V. Klein,
Consultant
View the video
Listen in MP3 format
View the presentation slides
CERT/CC was founded over two decades ago, SANS was created 20 years
ago, and USENIX has been providing a forum for world-class security
experts for even longer than that. Yet in spite of these prominent
centers of excellence (and their sage advice), we keep seeing new
attacks, new exploits, and new vulnerabilities—in simpler terms,
"same stuff, different day." It's not just because there are more
bad guys out there (although there are), and it's not just because
the bad guys are smarter (but they are). In my opinion it is because
we are working with tools and systems that are fundamentally flawed.
Our house of bricks is built on a sandy foundation, and we now find
ourselves at a crossroads—the same crossroads that every technology
has faced in human history: start over again and do it right from the
start, or keep doing it wrong until it all falls over in a heap.
This talk will try to take a lighthearted look at some really bad
news: either we will have to spend a lot of money redeveloping our
basic tools, infrastructure, and operating systems properly, or we
will have to spend a lot of money patching bugs and regularly
recovering from security disasters (and continually be faced with
the same basic problems). With one way we have a lot of unhappy people
now, while with the other we will have a lot of unhappy people later.
In the 1950s, the architect Frank Lloyd Wright was given a tour
of Pittsburgh which ended atop Mount Washington. He was asked, "What
should we do?" In his inimitable style, he looked around and said,
"Raze it and start over." Having lived in Pittsburgh for 35 years,
I can tell you that he was right.
I've worked with computers for as long as I've been in Pittsburgh.
Frank's advice is strangely apropos to my chosen profession, too . . .
|
|