COS-597B Syllabus

Schedule

This schedule and syllabus is preliminary and subject to change, particularly the second half of the semester. Papers marked by "everybody" should be read by -- surprise -- everybody in class. Papers not marked as such will be assigned through the review management software (HotCRP). See below for additional reviewing instructions.

Date Topics Readings Assignments
Wed 09/22 Course Overview (short class)
How to read papers (both Keshav and Mitzenmacher), End-to-end Argument, Hints for System Design - Everybody
Networking
Wed 09/29 Network architectures - Scalability and Configuration
FatTrees, VL2, Portland, SPAIN, Auto Addressing
Wed 10/06 Wide-area networking and datacenter TCP
Entact, Volley, TCP Incast, DC-TCP
Wed 10/13 Network architectures - High-throughput
BCube, CamCube, c-Through, Helios, Hedera
Virtualization and Management
Wed 10/20 Virtualization and Management
Difference Engine, Turtles, Fido, Eucalyptus, AutoControl
Wed 10/27 System and Lock Management
Paxos (Optional - Everybody); Chubby, AutoPilot, Zookeeper, Centrifuge Project groups chosen
Wed 11/03 Fall break and GEC9 (no class)
Storage and Data Processing
Wed 11/10 Storage (NoSQL)
Dynamo, PNUTS, CRAQ, TxCache, Stout Project pitches
Wed 11/17 Data Processing (MapReduce and next-gen DBMSs)
CACM Debate (MR v. DBs) - Everyone; MapReduce Online, Mantri, HStore, Deterministic DBs Project pitches
Wed 11/24 Data Processing and Storage (the Google way)
Piccolo, Nectar, BigTable, Percolator
Higher-level Topics
Wed 12/01 Trust and Security
Cloud Security, AVM, Depot, SPORC
Wed 12/08 Power and Green Computing
FAWN, FAWN Energy, LiteGreen, SleepServers, Sleepless in Seattle
Wed 12/15 Project presentations
Presentation (10-15 min)
Tue 01/11 Project papers
Papers due (11:59pm)

How to review

When reviewing papers, you will see a section for describing a paper summary, its strengths, its weaknesses, and detailed comments. In the summary section, please directly address:

  1. What problem the paper is addressing (1-2 sentences or bullets).
  2. The core novel ideas or technical contributions of the work (1-2 sentences or bullets). Put another way, what's the 30 second elevator pitch, or, five years from now, what should one remember about this paper?
  3. A longer description (3-5 sentences) that summarizes the paper's approach, mechanisms, and findings.

For the other sections, please include 2-4 bulletted points for the strengths and weaknesses, while a much longer exposition in the detailed comments. Remember to be constructive: don't only focus on the paper's shortcomings, but also on what it could have done differently or as the next steps. Imagine that you are having a conversation with the authors: What would you tell them?

In class, you might be called on to lead a discussion about the paper. Your oral description of the paper should follow a much similar format:

  1. System name (e.g., "VL2")
  2. Institution and/or authors (e.g., "Microsoft Research")
  3. Conference (e.g., "SIGCOMM 2009")
  4. Problem
  5. Core ideas
  6. Descriptive summary
  7. Strengths
  8. Weaknesses/limitations
  9. Further discussion, including proposals for follow-up work

Last updated: Wed Sep 12 20:22:43 -0400 2012