Advanced Topics in Computer Systems

Schedule & Reading List

We'll meet in for two hours every Thursday from 11.00—13.00 during Lent term starting January 23rd 2025. Each session after the first we will have three participant presentations which should be 15—20 minutes long.

We'll start with two presentations and a discussion followed by a short break, and then the final presentation and further discussion. The paper schedule for this year is given below, and here is the associated presentation schedule. Please be sure to check that you know which papers you are presenting, what flavour of presentation you should write, and in which slot (date and time) you are presenting.

Week 1: Reflections on Systems

Unlike the other sessions, this one is lecture format as you won't have had time to read any papers yet! The slides are now available here. The papers recommended in the lecture for reading are available below:

• How to read a paper, Keshav, S. 2007. ACM SIGCOMM CCR 37(3):83—84.
  [ DOI ]
• An evaluation of the ninth SOSP submissions or how (and how not) to write a good systems paper, Levin et al. 1983. ACM SIGOPS Operating Systems Review 17(3):35—40.
  [ DOI ]
• Writing reviews for systems conferences Roscoe, T. 2007.
  [ URL ]
• Hints for Computer Systems Design (Revised), Lampson, B. 1983. ACM SOSP 1983, pp.33—48.
  [ DOI (original) ]; [ URL (revised version) ]
• The role of motherhood in the pop art of system programming, Neumann, P. 1969. ACM SOSP 1969, pp.13—18.
  [ DOI ]
• Theory and practice in operating system design, Needham and Hartley. 1969. ACM SOSP 1969, pp.8—12.
  [ DOI ]

Week 2: Benchmarking

• MANET simulation studies: the incredibles, Kurkowski et al. 2005. ACM SIGMOBILE Mobile Computing and Communications Review, 9(4):50—61.
  [ DOI ]
• Open Versus Closed: A Cautionary Tale, Schroeder et al. 2006. USENIX NSDI 2006, pp.239—252.
  [ URL ]
• STABILIZER: statistically sound performance evaluation, Curtsinger et al. 2013. ACM ASPLOS 2013, pp.219—228.
  [ DOI ]

The following paper is only a short one (6pp) but is also well worth a read as it's highly readable, deeply insightful, and often quoted!

• Scalability! But at what COST?, McSherry et al. 2015. USENIX HOTOS 2015, pp.15—21.
  [ URL ]

Week 3: OS Structure

• Exokernel: an operating system architecture for application-level resource management, Engler et al. 1995. ACM SOSP 1995, pp.251—266.
  [ DOI ]
• Unikernels: Library Operating Systems for the Cloud, Madhavapeddy et al. 2013. ACM ASPLOS 2013, pp.461—472.
  [ DOI ]
• The Demikernel Datapath OS Architecture for Microsecond-scale Datacenter Systems, Zhang et al. 2021. ACM SOSP 2021, pp.195—211.
  [ DOI ]

Week 4: OS Abstractions

• Xen and the Art of Virtualization, Barham et al. 2003. ACM SOSP 2003, pp.164—177.
  [ DOI ]
• Light-Weight Contexts: An OS Abstraction for Safety and Performance, Litton et al. 2016. USENIX OSDI 2016, pp.49—64.
  [ URL ]
• Faster and Cheaper Serverless Computing on Harvested Resources, Zhang et al. 2021. ACM SOSP 2021, pp.724—739.
  [ DOI ]

Week 5: Distributed Consensus

• The Chubby lock service for loosely-coupled distributed systems, Burrows, M. 2006. USENIX OSDI 2006, pp.335—350.
  [ URL ]
• In Search of an Understandable Consensus Algorithm, Ongaro and Ousterhout. 2014. USENIX ATC 2014, pp.305—319.
  [ URL ]
• Basil: Breaking up BFT with ACID (transactions), Suri-Payer et al. 2021. ACM SOSP 2021, pp.1—17.
  [ DOI ]

Week 6: Cluster Scheduling

• Fuxi: a fault-tolerant resource management and job scheduling system at Internet scale, Zhang et al. 2014. Proc. VLDB Endowmet 7(13):1393—1404.
  [ DOI ]
• Large-scale cluster management at Google with Borg, Verma et al. 2015. ACM EuroSys 2015, pp.1—17.
  [ DOI ]
• Firmament: Fast, Centralized Cluster Scheduling at Scale, Gog et al. 2016. USENIX OSDI 2016, pp.99—115.
  [ URL ]

Week 7: Security & Privacy

• CryptDB: Protecting Confidentiality with Encrypted Query Processing, Popa et al. 2011. ACM SOSP 2011, pp.85—100.
  [ DOI ]
• Private Web Search with Tiptoe, Henzinger et al. 2023. ACM SOSP 2023, pp.396—416.
  [ DOI ]
• INVISILINE: Invisible Plausibly-Deniable Storage, Pinjala, S. K. et al. 2024. IEEE Symposium on Security & Privacy (SP) 2024, pp.2722—2739.
  [ DOI ]

Week 8: Side Channels

• Efficiently mitigating transient execution attacks using the unmapped speculation contract, Behrens et al. 2020. USENIX OSDI 2020, pp.1139—1154.
  [ URL ]
• Siloz: Leveraging DRAM Isolation Domains to Prevent Inter-VM Rowhammer, Loughlin et al. 2023. ACM SOSP 2023, pp.417—433.
  [ DOI ]
• AIR-FI: Leaking Data From Air-Gapped Computers Using Wi-Fi Frequencies, Guri, M. 2023. IEEE Transactions on Dependable & Secure Computing, 20(3):2547—2564
  [ DOI ]

Example presentations

To give you a feeling for the kinds of presentations expected, here are some more presentation examples. These are for papers not assigned this year, so you may find them a useful resource in any case. In some cases you will also find the conference presentations given by the original authors online too; while potentially useful, note that these serve a different purpose to and will usually be allocated more time than your presentations.

Advocate

• The Chubby lock service for loosely-coupled distributed systems, Burrows, M. 2006. USENIX OSDI 2006, pp.335—350.
  [ URL ]
• Xen and the Art of Virtualization, Barham et al. 2003. ACM SOSP 2003, pp.164—177.
  [ DOI ]
• ZooKeeper: Wait-free coordination for Internet-scale systems. Hunt et al, USENIX ATC 2010.
  [ URL ]
• My VM is Lighter (and Safer) than your Container Lupu et al, ACM SOSP 2017.
  [ DOI ]
• The multikernel: a new OS architecture for scalable multicore systems Baumann et al, ACM SOSP 2009.
  [ DOI ]

Balanced

• Open Versus Closed: A Cautionary Tale, Schroeder et al. 2006. USENIX NSDI 2006, pp.239—252.
  [ URL ]
• Unikernels: Library Operating Systems for the Cloud, Madhavapeddy et al. 2013. ACM ASPLOS 2013, pp.461—472.
  [ DOI ]
• Managing Update Conflicts in Bayou, a Weakly Connected Replicated Storage. Terry et al, ACM SOSP 1995.
  [ DOI ]
• Ryoan: A Distributed Sandbox for Untrusted Computation on Secret Data. Hunt et al, USENIX OSDI 2016.
  [ URL ]

Critical

• Firmament: Fast, Centralized Cluster Scheduling at Scale, Gog et al. 2016. USENIX OSDI 2016, pp.99—115.
  [ URL ]
• Faster and Cheaper Serverless Computing on Harvested Resources, Zhang et al. 2021. ACM SOSP 2021, pp.724—739.
  [ DOI ]
• Labels and Event Processes in the Asbestos Operating System. Efstathopolous et al, ACM SOSP 2005.
  [ DOI ]
• DryadLINQ: A System for General-Purpose Distributed Data-Parallel Computing Using a High-Level Language. Yu et al, USENIX OSDI 2008.
  [ URL ]
• Quincy: Fair Scheduling for Distributed Computing Clusters. Isard et al. ACM SOSP 2009.
  [ DOI ]
• Just say NO to Paxos Overhead: Replacing Consensus with Network Ordering. Li et al. USENIX OSDI 2016.
  [ URL ]
• MapReduce: Simplified Data Processing on Large Clusters. Dean and Ghemawat. USENIX OSDI 2004.
  [ URL ]
• MapReduce: Simplified Data Processing on Large Clusters. Dean and Ghemawat. USENIX OSDI 2004.
  [ URL ]
• GhostRider: A Hardware-Software System for Memory Trace Oblivious Computation. Liu et al, ACM ASPLOS 2015.
  [ DOI ]