Publications

Operating Systems: Principles and Practice, Second Edition

Thomas Anderson and Michael Dahlin
Recursive Books, 2014.

Sample schools using the book: Berkeley, Carnegie-Mellon, Cornell, ETH Zurich, Maryland, Michigan, NYU, Ohio, Penn State, Princeton, Stanford, Stony Brook, Texas-Austin Texas A&M, University of British Columbia, University of Central Florida, University of California Riverside, University of California Santa Barbara, University of Southern California, University of Washington

Strata: A Cross Layer File System

Peer to Peer Systems

• FreeDOM: a New Baseline for the Web
  Raymond Cheng, Will Scott, Arvind Krishnamurthy, and Thomas Anderson.
  HotNets 2012
  
An architecture for zero-configuration P2P web applications
• Scalable Consistency in Scatter
  Lisa Glendenning, Ivan Beschastnikh, Arvind Krishnamurthy, and Thomas Anderson
  SOSP 2011
  
A design and implementation of a scalable, consistent, and efficient distributed hash table
• Privacy-preserving P2P Data Sharing with OneSwarm
  Tomas Isdal, Michael Piatek, Arvind Krishnamurthy, and Thomas Anderson
  SIGCOMM 2010
  
Improving privacy without compromising file sharing performance
• Conference Reviewing Considered Harmful
  Thomas Anderson
  Operating Systems Review, April 2009
  
A game-theoretic analysis of the incentives induced by noisy conference reviewing
• Pitfalls for ISP-friendly P2P Design
  Michael Piatek, Harsha Madhyastha, John John, Arvind Krishnamurthy, and Thomas Anderson
  HotNets 2009
  Slides
  
Most BitTorrent swarms lack sufficient scale to benefit from ISP locality
• One Hop Reputations for Peer to Peer File Sharing Workloads
  Michael Piatek, Tomas Isdal, Arvind Krishnamurthy, and Thomas Anderson
  NSDI 2008
  
Incentives for sharing data without immediate reciprocation
• A Case for Holistic Incentive Design
  Michael Piatek, Thomas Anderson, and Arvind Krishnamurthy
  Workshop on Future Directions in Distributed Computing (FuDiCo) 2007
  
• Building BitTyrant, a (More) Strategic BitTorrent Client
  Michael Piatek, Tomas Isdal, T. Anderson, Arvind Krishnamurthy, and Arun Venkataramani
  ;login, August 2007
  
• Do Incentives Build Robustness in BitTorrent?
  Michael Piatek, Tomas Isdal, Thomas Anderson, Arvind Krishnamurthy, and Arun Venkataramani
  Best Student Paper, NSDI 2007
  
Contrary to popular belief, BitTorrent tit-for-tat can be gamed
• Amici Curiae, Metro-Goldwyn-Mayer Studios v. Grokster, Ltd
  Hal Abelson, Thomas Anderson, Andrew Appel, Steve Bellovin, Dan Boneh, David Clark, David Farber, Joan Feigenbaum, Ed Felten, Ron Harper, M. Frans Kaashoek, Brian Kernighan, Jennifer Rexford, John Reynolds, Avi Rubin, Eugene Spafford, and Dave Touretzky
  Submitted to the Supreme Court of the United States, February, 2005
  

Networking

• F10: Fault Tolerant Engineered Networks
  Vincent Liu, Dan Halperin, Arvind Krishnamurthy, and Thomas Anderson.
  NSDI 2013
  
Designing a data center network for sub-microsecond failure recovery.
• Tor Instead of IP
  Vincent Liu, Seungyeop Han, Arvind Krishnamurthy, and Thomas Anderson.
  
A proposal for a censorship resistant Internet.
• ETTM: A Scalable Fault Tolerant Network Manager
  Colin Dixon, Hardeep Uppal, Vjekoslav Brajkovic, Dane Brandon, Thomas Anderson, and Arvind Krishnamurthy
  NSDI 2011
  Video
  
Flexible enterprise network management services running in a secure partition on end hosts
• End to the Middle
  Colin Dixon, Thomas Anderson and Arvind Krishnamurthy
  HotOS 2009
  
• Consensus Routing: The Internet as a Distributed System
  John John, Ethan Katz-Bassett, Arvind Krishnamurthy, Thomas Anderson, and Arun Venkataramani
  Best Paper, NSDI 2008
  
Protocol to mask almost all outages due to BGP convergence
• Achieving Convergence-Free Routing Using Failure-Carrying Packets
  Karthik Lakshminarayanan, Matthew Caesar, Murali Rangan, Thomas Anderson, Scott Shenker, and Ion Stoica
  SIGCOMM 2007
  
Dynamically re-route packets that would otherwise traverse a failed link
• Mutually Controlled Routing with Independent ISPs
  Ratul Mahajan, David Wetherall, and Thomas Anderson
  NSDI 2007
  
A protocol for ISPs to trade efficient routes with each other
• PCP: Efficient Endpoint Congestion Control
  Thomas Anderson, Andy Collins, Arvind Krishnamurthy, and John Zahorjan
  NSDI 2006
  
Low latency, zero loss, high throughput TCP congestion control
• GENI Conceptual Design, Project Execution Plan
  Thomas Anderson, Daniel Blumenthal, Dean Casey, David Clark, Deborah Estrin, Larry Peterson, Dipankar Raychaudhry, Jennifer Rexford and John Wroclawski
  NSF Major Research Equipment Facility Construction Proposal, January 2006
  
A case study of why to keep projects as small as feasible
• Overcoming Barriers to Disruptive Innovation in Networking
  Thomas Anderson, Larry Peterson, Scott Shenker and Jon Turner, editors
  Report of NSF Workshop, January 2005
  
Advocates that NSF focus on long-range and experimental networking research
• Negotiation-Based Routing Between Neighboring ISPs
  Ratul Mahajan, David Wetherall, and Thomas Anderson
  NSDI 2005
  
A protocol for neighboring ISPs to trade efficient routes with each other
• Overcoming the Internet Impasse through Virtualization
  Thomas Anderson, Larry Peterson, Scott Shenker and Jon Turner
  IEEE Computer, April 2005
  
A case for reinventing the Internet by building an experimental platform for evaluating new ideas
• Towards Coordinated Interdomain Traffic Engineering
  Ratul Mahajan, David Wetherall, and Thomas Anderson
  HotNets 2004
  
• On the Stability of Adaptive Routing in the Presence of Congestion Control
  Eric Anderson and Thomas Anderson
  IEEE Infocom 2003
  
Routing and congestion control can be decoupled, if the traffic matrix is admissible
• Design Considerations for Robust Internet Protocols
  Thomas Anderson, Scott Shenker, Ion Stoica, and David Wetherall
  HotNets 2002
  
Case studies of security and robustness flaws in the design of network protocols
• Robust Congestion Signalling
  David Ely, Neil Spring, David Wetherall, Stefan Savage, and Thomas Anderson
  IEEE International Conference on Network Protocols (ICNP) 2001
  
Prevent intermediaries from spoofing ECN using a randomized nonce bit
• Network Support for IP Traceback
  Stefan Savage, David Wetherall, Anna Karlin, and Thomas Anderson
  IEEE/ACM Transactions on Networking, June 2001; SIGCOMM 2000
  
A mechanism to identify the source of a denial-of-service attack
• Modeling TCP Latency
  Neal Cardwell, Stefan Savage, and Thomas Anderson
  IEEE Infocom 2000
  
Model of TCP startup behavior in the presence of loss
• Understanding the Performance of TCP Pacing
  Amit Aggarwal, Stefan Savage, and Thomas Anderson
  IEEE Infocom 2000
  
Spacing out TCP packets can have counterintuitive performance effects
• Receiver Based Management of Low Bandwidth Access Links
  Neil Spring, Maureen Chesire, Mark Berryman, Vivek Sahasranaman, Thomas Anderson, and Brian Bershad
  IEEE Infocom 2000
  
Receiver driven packet shaping
• TCP Congestion Control with a Misbehaving Receiver
  Stefan Savage, Neal Cardwell, David Wetherall and Thomas Anderson
  ACM Computer Communications Review (CCR), October 1999
  
A TCP receiver can spoof a sender to ignore congestion signals from the network
• The Case for Informed Transport Protocols
  Stefan Savage, Neal Cardwell and Thomas Anderson
  HotOS 1999
  
Argues for using history for speeding TCP startup; helps motivate PCP
• Detour: Informed Internet Routing and Transport
  Stefan Savage, Thomas Anderson, Amit Aggarwal, David Becker, Neal Cardwell, Andy Collins, Eric Hoffman, John Snell, Amin Vahdat, Geoff Voelker, and John Zahorjan
  IEEE Micro, February 1999; Award Paper, Hot Interconnects, 1998
  
This paper was the first to identify the problem of pathological Internet routing and to propose the use of overlay routing to compensate for the poor behavior of the Internet. Commercial versions of detour routing are available from Akamai and Limelight.
• A Quantitative Comparison of Scheduling Algorithms for Input-Queued Switches
  Nick McKeown and Thomas Anderson
  Computer Networks and ISDN Systems, 1998
• Modeling and Optimizing Pipeline Latency
  Randy Wang, Arvind Krishnamurthy, Rich Martin, Thomas Anderson, and David Culler
  SIGMETRICS 1998
  
It is possible to simulate cut-through on a store-and-forward network
• High Speed Switch Scheduling for Local Area Networks
  Thomas Anderson, Susan Owicki, James Saxe, and Charles Thacker
  ACM Transactions on Computer Systems, November 1993, Award Paper, ASPLOS 1992
  
A very high-speed crossbar scheduling algorithm, for switches without head of line blocking

Network Measurement and Diagnosis

• LIFEGUARD: Practical Repair of Persistent Route Failures
  Ethan Katz-Bassett, Colin Scott, David Choffnes, Italo Cunha, Vytautas Valancius, Nick Feamster, Harsha Madhyastha, Thomas Anderson, and Arvind Krishnamurthy. SIGCOMM 2012
  
For long-lasting, partial Internet outages, an architecture and implementation for how the affected end hosts can induce the rest of the network to repair itself.
• Resolving IP Aliases with Prespecified Timestamps
  Justine Sherry, Ethan Katz-Bassett, Mary Pimenova, Harsha Madhyastha, Arvind Krishnamurthy, And Thomas Anderson
  IMC 2010
  
A paper showing it is possible to work around the difficulties of using IP Timestamps
• Reverse Traceroute
  Ethan Katz-Bassett, Harsha Madhyastha, Vijay Adhikari, Colin Scott, Justine Sherry, Peter van Wesep, Thomas Anderson and Arvind Krishnamurthy
  Best Paper, NSDI 2010
  Slides and Video
  
A tool to measure the reverse path back from any Internet host
• Moving Beyond End-to-End Path Information to Optimize CDN Performance
  Rupa Krishnan, Harsha Madhyastha, Sushant Jain, Sridhar Srinivasan, Arvind Krishnamurthy, Thomas Anderson, and Jie Gao
  Best Paper, IMC 2009
  
There are many problems on paths from client machines to data centers
• iPlane Nano: Path Prediction for Peer-to-Peer Applications
  Harsha Madhyastha, Ethan Katz-Bassett, Thomas Anderson, Arvind Krishnamurthy, and Arun Venkataramani
  NSDI 2009
  
Compact representation of the Internet topology and path characteristics
• Studying Blackholes in the Internet with Hubble
  Ethan Katz-Bassett, Harsha Madhyastha, John John, Arvind Krishnamurthy, David Wetherall, and Thomas Anderson
  NSDI 2008
  
Partial outages are much more common than previously thought
• Profiling a Million User DHT
  Jarret Falkner, Michael Piatek, John John, Arvind Krishnamurthy, and Thomas Anderson
  IMC 2007
  
Kademlia is slow, inconsistent and unreliable
• Leveraging BitTorrent for End Host Measurements
  Tomas Isdal, Michael Piatek, Arvind Krishnamurthy, and Thomas Anderson
  Passive and Active Measurement Conference (PAM) 2007
  
Estimate uplink bandwidth from spacing of BitTorrent packet pairs
• iPlane: An Information Plane for Distributed Services
  Harsha Madhyastha, Tomas Isdal, Michael Piatek, Colin Dixon, Thomas Anderson, Arvind Krishnamurthy, and Arun Venkataramani
  OSDI 2006
  
A comprehensive measurement of the Internet topology is possible, and useful
• Towards IP Geolocation using Delay and Topology Measurements
  Ethan Katz-Bassett, John John, Arvind Krishnamurthy, David Wetherall, and Thomas Anderson
  IMC 2006
Use topology and delay to constrain the location of routers and hosts
• A Structural Approach to Latency Prediction
  Harsha Madhyastha, Thomas Anderson, Arvind Krishnamurthy, Neil Spring, and Arun Venkataramani
  IMC 2006
• Reverse-Engineering the Internet
  Neil Spring, David Wetherall, and Thomas Anderson
  HotNets, 2002
  
• User-level Internet Path Diagnosis
  Ratul Mahajan, Neil Spring, David Wetherall and Thomas Anderson SOSP 2003
  
An architecture for diagnosing problems in the Internet from end hosts
• Measuring ISP Topologies with Rocketfuel
  Neil Spring, Ratul Mahajan, David Wetherall, and Thomas Anderson
  IEEE/ACM Transactions on Networking February 2004
  2005 IEEE Communications Society William R. Bennett Prize
  
It is possible to measure the topology of a single ISP
• Quantifying the Causes of Path Inflation
  Neil Spring, Ratul Mahajan, and Thomas Anderson
  SIGCOMM 2003
  
Path inflation is primarily due to BGP and peering policy
• Scriptroute: A Facility for Distributed Internet Measurement
  Neil Spring, David Wetherall, and Thomas Anderson
  Award paper, USITS 2003
  
A flexible but safe Internet measurement platform
• Inferring Link Weights Using End-to-End Measurements
  Ratul Mahajan, Neil Spring, David Wetherall, and Thomas Anderson
  SIGCOMM Internet Measurement Workshop (IMW) 2002
  
It is possible to infer routing policy using distributed traceroutes
• Understanding BGP Misconfiguration
  Ratul Mahajan, David Wetherall, and Thomas Anderson
  SIGCOMM 2002
  
Errors in configuring BGP are widespread
• The End-to-End Effects of Internet Path Selection
  Stefan Savage, Andy Collins, Eric Hoffman, John Snell, and Thomas Anderson
  SIGCOMM 1999
  
Many Internet routes violate the triangle inequality

Security

• Retaining Sandbox Containment Despite Bugs in Privileged Memory-safe Code
  Justin Cappos, Armon Dadgar, Jeff Rasley, Justin Samuel, Ivan Beschastnikh, Cosmin Barsan, Arvind Krishnamurthy, and Thomas Anderson
  CCS 2010
  
• TVA: A DoS-Limiting Network Architecture
  Xiaowei Yang, David Wetherall, and Thomas Anderson
  IEEE/ACM Transactions on Networking, December 2008
  
Network capabilities to provide receivers control over who can send to them
• Withstanding Multimillion-node Botnets
  Colin Dixon, Thomas Anderson, and Arvind Krishnamurthy
  ;login, August 2008
  
• Phalanx: Withstanding Multimillion-Node Botnets
  Colin Dixon, Thomas Anderson, and Arvind Krishnamurthy
  NSDI 2008
  
Foiling DoS using randomized intermediary mailboxes
• Efficient and Secure Source Authentication with Packet Passports
  Xin Liu, Xiaowei Yang, David Wetherall, and Thomas Anderson
  Workshop on Steps to Reducing Unwanted Traffic on the Internet (SRUTI) 2006
  
• GENI Facility Security
  Thomas Anderson and Michael Reiter
  GENI Design Document 06-23, September 2006
  
How to put PlanetLab-style systems on a more secure foundation
• A DoS-Limiting Network Architecture
  Xiaowei Yang, David Wetherall, and Thomas Anderson
  SIGCOMM 2005
  
• Preventing Internet Denial-of-Service with Capabilities
  Thomas Anderson, Timothy Roscoe, and David Wetherall
  HotNets 2003
  
Proposal for network capabilities as a mechanism to give receivers control over who is allowed to send to them
• CRISIS: A Wide Area Security Architecture
  Eshwar Belani, Amin Vahdat, Thomas Anderson, and Michael Dahlin
  USENIX Security 1998
  
A proposed simplification of the Taos security architecture for cloud services

Wireless and Pervasive Systems

• Taking the Sting out of Carrier Sense: Interference Cancellation for Wireless LANs
  Dan Halperin, Thomas Anderson, and David Wetherall
  Mobicom 2008 2008
  
An experimental evaluation of interference cancellation
• Interference Cancellation: Better Receivers for a New Wireless MAC
  Dan Halperin, Josephine Ammer, Thomas Anderson, and David Wetherall
  HotNets 2007
  
Proposes a new algorithm for interference cancellation for WiFi networks
• System Support for Pervasive Applications
  Robert Grimm, Janet Davis, Eric Lemar, Adam MacBeth, Steve Swanson, Thomas Anderson, Brian Bershad, Gaetano Borriello, Steve Gribble, and David Wetherall
  ACM Transactions on Computer Systems November 2004.
  
Service discovery and other API support for mobile applications
• System-Level Programming Abstractions for Ubiquitous Computing
  Robert Grimm, Janet Davis, Eric Lemar, Adam MacBeth, Steve Swanson, Steve Gribble, Thomas Anderson, Brian Bershad, Gaetano Borriello, and David Wetherall
  Proc. UbiTools '01 Workshop 2001
• System Directions for Pervasive Computing
  Robert Grimm, Janet Davis, Ben Hendrickson, Eric Lemar, Adam MacBeth, Steve Swanson, Thomas Anderson, Brian Bershad, Gaetano Borriello, Steve Gribble, and David Wetherall
  HotOS 2001
  
• A System Architecture for Pervasive Computing
  Robert Grimm, Thomas Anderson, Brian Bershad, David Wetherall
  ACM SIGOPS European Workshop 2000
  
• Next Century Challenges: Data-Centric Networking for Invisible Computing
  Mike Esler, Jeffrey Hightower, Thomas Anderson and Gaetano Borriello
  MOBICOM 1999.

Cloud Computing

• Learning from PlanetLab
  Thomas Anderson and Timothy Roscoe
  Workshop on Real, Large Distributed Systems (WORLDS) 2006
  
A post-mortem: what we might have done differently, with the benefit of hindsight
• GENI Distributed Services
  Thomas Anderson and Amin Vahdat (Ed.)
  GENI Design Document 06-24, September 2006
  
A set of ideas for fostering experimental distributed systems research
• A Blueprint for Introducing Disruptive Change in the Internet
  Larry Peterson, Thomas Anderson, David Culler, Timothy Roscoe
  HotNets 2002
  
Introduces PlanetLab as a platform for fostering Internet innovation
• Active Names: Flexible Location and Transport of Wide-Area Resources
  Amin Vahdat, Michael Dahlin, Thomas Anderson and Amit Aggarwal
  USENIX Symposium on Internet Technologies and Systems (USITS) 1999
  
Programmable control of name translation for flexible service delivery
• WebOS: Operating System Services for Wide Area Applications
  Amin Vahdat, Thomas Anderson, Michael Dahlin, Eshwar Belani, David Culler, Paul Eastham, and Chad Yoshikawa
  High Performance Distributed Computing (HPDC) 1998
  
A very early version of what later became known as cloud computing
• GLUnix: A Global Layer UNIX for a Network of Workstations
  Doug Ghormley, David Petrou, Steve Rodrigues, Amin Vahdat, and Thomas Anderson
  Software Practice and Experience, July 1998
  
Provide a single system UNIX API over a cluster
• Effects of Communication Latency, Overhead and Bandwidth in a Cluster Architecture
  Rich Martin, Amin Vahdat, David Culler and Thomas Anderson
  ISCA 1997
  
Evaluates sensitivity of a set of parallel applications to message latency, overhead and bandwidth
• High-Performance Local Area Communication with Fast Sockets
  Steve Rodrigues, Thomas Anderson, and David Culler
  USENIX Conference, January 1997
  
Extends sockets to provide low overhead LAN communication
• Using Smart Clients to Build Scalable Services
  Chad Yoshikawa, Brent Chun, Paul Eastham, Amin Vahdat, Thomas Anderson, and David Culler
  USENIX Conference January 1997
  
Client-side flexibility is helpful in managing distributed cloud services
• WebFS: A Global Cache Coherent Filesystem
  Amin Vahdat, Paul Eastham, and Thomas Anderson
  UC Berkeley Technical Report, December 1996
  
• LogP Quantified: The Case for Low Overhead Local Area Networks
  Kim Keeton, Thomas Anderson, and David Patterson
  Hot Interconnects 1995
  
• The Interaction of Parallel and Sequential Workloads on a Network of Workstations
  Remzi Arpaci, Andrea Dusseau, Amin Vahdat, Lok Liu, Thomas Anderson, and David Patterson
  SIGMETRICS 1995
  
• A Case for NOW (Networks of Workstations)
  Thomas Anderson, David Culler, David Patterson, and the NOW Team
  IEEE Micro, February 1995, Award Paper Hot Interconnects 1994
  
This paper makes the case for the (then controversial, now standard) idea of building scalable services out of clusters of commodity computers. A spinoff of this project was the search engine pioneer Inktomi, acquired by Yahoo! Among the innovations described in the paper are the use of cluster memory to improve file system performance and providing a unified scalable network storage abstraction across the cluster.

File Systems

• Trading Capacity for Performance in a Disk Array
  Xiang Yu, Benjamin Gum, Yuqun Chen, Randy Wang, Arvind Krishnamurthy, Kai Li, and Thomas Anderson
  OSDI 2000
  
Staggers mirrored data across an array to reduce both seek and latency delay
• A Comparison of File System Workloads
  Drew Roselli, Jay Lorch, and Thomas Anderson
  USENIX Conference, June 2000
  
• Virtual Log-Based File Systems for a Programmable Disk
  Randy Wang, Thomas Anderson, and David Patterson
  OSDI 1999
  How to achieve very low latency disk writes
• Experience with a Distributed File System Implementation
  Randy Wang, Thomas Anderson, and Michael Dahlin
  Unpublished manuscript, 1998
  
A post-mortem for xFS: we should have built it at user-level
• Improving the Performance of Log Structured File Systems with Adaptive Methods
  Jeanna Neefe Matthews, Drew Roselli, Adam Costello, Randy Wang, and Thomas Anderson
  SOSP 1997
  
Self-tuning xFS parameters
• Serverless Network File Systems
  Thomas Anderson, Michael Dahlin, Jeanna Neefe, Drew Roselli, David Patterson, and Randy Wang
  ACM Transactions on Computer Systems, February 1996, Award Paper, SOSP 1995
  
For scalability, design the filesystem as you would a parallel program
• Cooperative Caching: Using Remote Client Memory to Improve File System Performance
  Michael Dahlin, Thomas Anderson, David Patterson, and Randy Wang
  OSDI, 1994
  
Use memory in machines over a LAN as a very large scale distributed file cache
• A Quantitative Analysis of Cache Policies for Scalable File Systems
  Michael Dahlin, Thomas Anderson, Cliff Mather, David Patterson and Randy Wang
  SIGMETRICS 1994
  
An evaluation of a peer to peer distributed file system
• A Quantitative Analysis of Disk Drive Power Management in Portable Computers
  Kester Li, Roger Kumpf, Paul Horton, and Thomas Anderson
  Winter USENIX Conference, January 1994
  
Spinning down the disk saves energy, but makes reads/writes slow
• xFS: A Wide Area Mass Storage File System
  Randy Wang and Thomas Anderson
  Workshop on Workstation Operating Systems (WWOS), 1993
  
Proposes write-back cache coherence as a design principle for wide area file systems

Computer Architecture

• Execution Characteristics of Desktop Applications on Windows NT
  Dennis Lee, Patrick Crowley, Jean-Loup Baer, Thomas Anderson, and Brian Bershad
  ISCA 1998
  
• Scalable Processors to 1 Billion Transistors and Beyond: IRAM
  Christoforos Kozyrakis, Stylianos Perissakis, David Patterson, Thomas Anderson, Krste Asanovic, Neal Cardwell, Richard Fromm, Jason Golbus, Benjamin Gribstad, Kimberly Keeton, Randi Thomas, Noah Treuhaft, and Katherine Yelick
  IEEE Computer, September 1997
  
• The Energy Efficiency of IRAM Architectures
  Richard Fromm, Stylianos Perissakis, Neal Cardwell, Christoforos Kozyrakis, Bruce McGaughy, Dave Patterson, Thomas Anderson, and Kathy Yelick
  ISCA 1997
  
Evaluates the performance/energy tradeoff of combining DRAM and computation onto the same chip
• A Case for Intelligent RAM: IRAM
  David Patterson, Thomas Anderson, Neal Cardwell, Richard Fromm, Kim Keeton, Christoforos Kozyrakis, Randi Thomas, and Kathy Yelick
  IEEE Micro, April 1997, Award Paper, Hot Chips 1996
  
Proposes combining DRAM and computation onto the same chip
• Intelligent RAM (IRAM): Chips That Remember and Compute
  David Patterson, Thomas Anderson, Neal Cardwell, Richard Fromm, Kim Keeton, Christoforos Kozyrakis, Randi Thomas, and Kathy Yelick
  IEEE International Solid-State Circuits Conference (ISSCC) 1997
  
• The Interaction of Architecture and Operating System Design
  Thomas Anderson, Hank Levy, Brian Bershad, and Ed Lazowska ASPLOS 1991
  
On modern processors, the performance of operating system functions does not improve at anywhere near the same rate as application code

Operating Systems

• Reducing Startup Latency in Web and Desktop Applications
  Dennis Lee, Jean-Loup Baer, Brian Bershad, and Thomas Anderson
  USENIX Windows NT Symposium 1999
  
• Transparent Result Caching
  Amin Vahdat and Thomas Anderson
  USENIX Conference, June 1998
  
Automates provenance tracking, so that rebuilds happen automatically when data files or tool chains change
• SLIC: An Extensibility System for Commodity Operating Systems
  Doug Ghormley, David Petrou, Steve Rodrigues, and Thomas Anderson
  USENIX Conference, June 1998
  
A system for inserting code at the system call interface
• Thread Management for Shared-Memory Multiprocessors
  Thomas Anderson, Brian Bershad, Ed Lazowska, and Hank Levy
  The Computer Science and Engineering Handbook, Allen Tucker, Ed., CRC Press, 1997
  
• Tools for the Development of Application-Specific Virtual Memory Management
  Keith Krueger, David Loftesness, Amin Vahdat, and Thomas Anderson
  OOPSLA 1993
  
• Efficient Software-Based Fault Isolation
  Robert Wahbe, Steve Lucco, Thomas Anderson and Susan Graham
  SOSP 1993
  
This paper was the first to provide software sandboxing for the safe execution of third party plug-ins, written in any arbitrary language. Software sandboxing is widely used today in web browsers; the language independent approach taken by this paper is being used in Google's Native Client system for executing x86 plug-ins. It has also been proposed for sandboxing kernel device drivers, the primary source of operating system crashes.
• Anonymous RPC: Low Latency Protection in a 64-Bit Address Space
  Curtis Yarvin, Richard Bukowski, and Thomas Anderson
  Best Student Paper, Summer USENIX Conference, June 1993
  
• The Case for Application-Specific Operating Systems
  Thomas Anderson
  Workshop on Workstation Operating Systems (WWOS) 1992
  
Applications can do a better job of optimizing their resources than the operating system can on their behalf
• Scheduler Activations: Effective Kernel Support for the User-Level Management of Parallelism
  Thomas Anderson, Brian Bershad, Ed Lazowska and Hank Levy
  ACM Transactions on Computer Systems, February 1992
  Award Paper, SOSP 1991
  
Operating system kernel support for efficient user-level threads. Scheduler activations have been implemented in several research operating systems (e.g., MIT's Exokernel and MSR's Barrelfish) and were most recently were implemented in Windows 7 as part of Microsoft’s strategy to support multicore architectures.
• Operating System Support for High Performance Multiprocessing
  Thomas Anderson
  Ph.D. Thesis, University of Washington, UW Technical Report 91-08-10, August 1991
  
• User-Level Interprocess Communication for Shared-Memory Multiprocessors
  Brian Bershad, Thomas Anderson, Ed Lazowska and Hank Levy
  ACM Transactions on Computer Systems, May 1991
  
On a multiprocessor, very fast interprocess communication can be provided through shared memory portals
• Lightweight Remote Procedure Call
  Brian Bershad, Thomas Anderson, Ed Lazowska and Hank Levy
  ACM Transactions on Computer Systems, February 1990
  Award Paper, SOSP 1989
  
On a uniprocessor, very fast interprocess communication can be provided through a careful kernel implementation
• The Performance of Spin Lock Alternatives for Shared-Memory Multiprocessors
  Thomas Anderson
  IEEE Transactions on Parallel and Distributed Systems, January 1990
  International Conference on Parallel Processing (ICPP) 1989
  
Contended spinlocks do not perform well with cache coherence; proposes queueing and exponential backoff as ways of achieving better scalability
• The Performance Implications of Thread Management Alternatives for Shared-Memory Multiprocessors
  Thomas Anderson, Ed Lazowska, and Hank Levy
  IEEE Transactions on Computers, December 1989
  Award Paper, SIGMETRICS 1989
  
A very fast user-level thread package

Software Engineering

• Eraser: A Dynamic Data Race Detector for Multi-Threaded Programs
  Stefan Savage, Mike Burrows, Greg Nelson, Patrick Sobalvarro, and Thomas Anderson
  ACM Transactions on Computer Systems, November 1997
  Award Paper, SOSP 1997
  
This paper invented the lock set algorithm, the first practical approach to automatically identifying data races in multithreaded programs using mutual exclusion for synchronization. This algorithm is in wide use in toolkits for writing and debugging multithreaded programs.
• Experience with a Language for Writing Coherence Protocols
  Satish Chandra, Mike Dahlin, Brad Richards, Randy Wang, Thomas Anderson, and Jim Larus
  USENIX Conference on Domain-Specific Languages, October 1997
  
• Tuning Memory Performance in Sequential and Parallel Programs
  Margaret Martonosi, Anoop Gupta, and Thomas Anderson
  IEEE Computer, April 1995
  
• Effectiveness of Trace Sampling for Performance Debugging Tools
  Margaret Martonosi, Anoop Gupta, and Thomas Anderson
  SIGMETRICS 1993
  
How to use sampling to improve the performance of cache simulations
• MemSpy: Analyzing Memory System Bottlenecks in Programs
  Margaret Martonosi, Anoop Gupta, and Thomas Anderson
  SIGMETRICS 1992
  
A tool for debugging the cache behavior of application code
• Quartz: A Tool for Tuning Parallel Program Performance
  Thomas Anderson and Ed Lazowska
  SIGMETRICS 1990
  
A tool for identifying performance bottlenecks in parallel code, by inflating sampled CPU cycles by the sampled amount of concurrency

Education

• Seattle: A Platform for Educational Cloud Computing
  Justin Cappos, Ivan Beschastnikh, Arvind Krishnamurthy, and Thomas Anderson
  SIGCSE 2009
• The Nachos System
  Thomas Anderson
  Appendix in Operating Systems Concepts, 4th Ed., A. Silberschatz and P. Galvin
  
• The Nachos Instructional Operating System
  Wayne Christopher, Steve Procter, and Thomas Anderson
  Best Paper, Winter USENIX Conference, January 1993
  
A widely used project for teaching undergraduate operating systems