CPS 510 (Duke University) Advanced Operating Systems

Here are the topics and readings for the semester. The 12 papers with bold tags are assigned reading. I might make a few minor adjustments, but this is the plan. I have also listed some relevant "classics" that we will touch on in the discussions.

Rethinking operating system structure

[Singularity] Singularity: Rethinking the Software Stack. By Galen Hunt and Jim Larus. ACM SIGOPS Operating Systems Review 41,2 (April 2007). [local pdf]. Skip section 4 (except 4.2).
Singularity project at Microsoft Research (MSR)
Author slides on Singularity (ASPLOS 08) [local pdf]
[Drawbridge] Rethinking the Library OS from the Top Down By Donald Porter, Silas Boyd-Wickizer, Jon Howell, Reuben Olinsky, and Galen Hunt. In ASPLOS, March 2011. [local pdf]
Drawbridge project at MSR

Optional classics:

Virtual Memory Primitives for User Programs. By Appel and Li. In ASPLOS, April 1991. [local pdf]
Extensibility, Safety and Performance in the SPIN Operating System. By Bershad et al. In SOSP, Dec 1995. [local pdf]
Exokernel: an OS Architecture for Application-level Resource Management. Engler et al. 15th SOSP, Dec 1995. [local pdf]

Concurrency and communication

[Threads] An Introduction to Programming with C# Threads (2003). By Andrew Birrell. This paper is a revision of the classic introduction to threads. [local pdf]
[Memory] The Java Memory Model, Jeremy Manson, William Pugh, Sarita V. Adve, Proceedings of the 32nd Symposium on Principles of Programming Languages (POPL) Jan. 2005. [local pdf]. Read Sections 1-4.
See also The Java Memory Model from Java Concurrency in Practice
[RPC] Performance of Firefly RPC. By Schroeder and Burrows. ACM TOCS 8(1), 1990 (from SOSP 1989). [local pdf]

Optional classics to discuss:

Time, Clocks, and the Ordering of Events in a Distributed System. By Leslie Lamport. Commun. ACM, 21(7):558-565, 1978. [local pdf]
Scheduler Activations: Effective Kernel Support for the User-level Management of Parallelism. In the Thirteenth ACM Symposium on Operating Systems Principles (SOSP), December 1991. By Tom Anderson with Bershad, Lazowska, Levy. [local pdf]

Architectural foundations

[Dune] Dune: Safe User-level Access to Privileged CPU Features By Belay et. al. In Proceedings of the 10th Symposium on Operating Systems Design and Implementation (OSDI), Hollywood, CA, October 2012. [local pdf]
[Haven] Shielding Applications from an Untrusted Cloud with Haven. By Andrew Baumann, Marcus Peinado, and Galen Hunt. In the 11th USENIX Symposium on Operating Systems Design and Implementation (OSDI), October 2014. [local pdf]

Optional classics:

The Interaction of Architecture and Operating System Design . Anderson, Levy, Bershad, Lazowska. ASPLOS 1991.
The Multics Virtual Memory: Concepts and Design (1972).
A Survey of Virtual Machine Research (1974). By Robert P. Goldberg.

Access control

[Flume] Information Flow Control for Standard OS Abstractions. By Krohn et. al., SOSP 2007. [local pdf]

Optional classics:

A Lattice Model of Secure Information Flow. By Dorothy Denning (1976). [local pdf]

Recovery and replication

[Remus] Remus: High Availability Via Asynchronous Virtual Machine Replication. By Brendan Cully, Geoffrey Lefebvre, Dutch Meyer, Mike Feeley, Norm Hutchinson, and Andrew Warfield. In NSDI'08: Proceedings of the 5th USENIX Symposium on Networked Systems Design and Implementation, April 2008. [local pdf]
Optional classics: TBD

Storage and scalability

[GFS] The Google File System. By Sanjay Ghemawat, Howard Gobioff, and Shun-Tak Leung. In the Proceedings of the ACM Symposium on Operating Systems Principles (SOSP), October 2003. [local pdf]
Authors' GFS slides [PDF]
[DDS] Scalable Distributed Data Structures for Internet Service Construction . By Steve Gribble with Brewer, Hellerstein, and Culler. Proceedings of OSDI 2000. [local pdf]
Murat's blog entry on the DDS paper
[Spanner] Spanner: Google's Globally Distributed Database. By Corbett et. al. Proceedings of OSDI 2012. [local pdf]
Replicated Data Consistency Explained Through Baseball. By Doug Terry. (2011)

Optional classics:

Leases: an efficient fault-tolerant mechanism for distributed file cache consistency. By Gray and Cheriton. In SOSP 1989. [local pdf]
Disconnected operation in the Coda File System By Kistler and Satyanarayanan. ACM TOCS 10(1), Feb 1992.