COS-518 Semester Project

The semester-long project in COS-518 is an open-ended systems research project, interpreted broadly. The projects may be proposed independently, or they may be drawn from a list of suggested topics. Projects should be done individually or groups of two, with the approval of the instructor.

The project time-line is as follows:

Proposal: Project proposals are due on 10/18. These should be a few pages that succinctly describe what you plan to work on, a rough time-line, and a bit of the related work in the area.
Proposal discussion: One-on-one meeting with instructor to discuss projects during week of October 22nd. Meetings to be scheduled.
Project status discussion: One-on-one meeting with instructor to discuss projects during week of December 3rd. Meetings to be scheduled.
Final presentation: Groups are expected to present a high-quality 15-minute presentation of their project, similar in structure and scope to conference presentations. These presentations will take place during on 1/11.
Final report: Project writeups are due 1/15 at 11:59pm EST (emailed .pdf and tar ball of all source code). Project writeups should be 10 pages of double-columned, single-spaced, eleven-point font. Use this skeleton framework for formatting and building your paper.
Reviews: Finally, after project writeups are submitted, each individual will be assigned to review the project reports of one to two other groups. Reviews are expected to be at least one page (typed) per project, similar in scope to your paper critiques (and especially focusing on the project's strengths, weaknesses, and areas for future consideration). This task is similar to that which is performed by program committee (PC) members of conferences. Assigned reviews of peers' projects due 1/20 at 11:59pm EDT (emailed text).

One project will be selected for a best project award, based on paper writeups and presentations. The decision will be based on both class participants' votes and the instructor's discretion.

Project Suggestions

List of possible topics can be found here. To access the page, you either need to be using a computer on the Princeton network or know the password for remote access.

Project Presentation Schedule for Jan 11

Group	Title	Time
Golightly / W. Kim	File Synchronization for Iteractive Use	11:00 am
Chen / Huang	CACIS: Content-Aware Cooperative Image Storage	11:30 am
Sevinc / Yu	CPU Allocation for Packet Forwarding of Virtual Networks on Commodity Hardware	11:55 am
Green / Keller	Shared Kernel-Mode Click for Programmable Virtual Networks	12:20 pm
Kumar / Koukoumidis	u-Multicast: Lightweight Ad-hoc Multicast for Mobile Internet Applications	1:15 pm
Jung / H. Kim	Scheduler for Parallelized Blocks	1:40 pm
Lloyd / Sen	A Fault-Tolerance Abstraction for Scalable Byzantine Systems	2:05 pm
Terrace	A Dynamically Reliable DHT Cluster	2:30 pm
Badam / Ihm	10g -- A Super-Saver WAN Accelerator (Best Project)	2:55 pm

Project Management

You should strongly consider using either Subversion or CVS to perform source code control for your project and the paper you write describing it. I suggest Subversion.

I also strongly suggest writing your course project report using LaTeX. It is the de-facto tool in which most CS research papers are written. While it has a bit of start up cost, it's much easier to collaboratively write complex research papers using LaTeX than using word.

Writing Papers

Sample LaTeX paper (tar.gz)
An Evaluation of the Ninth SOSP Submissions - or - How (and How Not) to Write a Good Systems Paper by Roy Levin and David D. Redell
Some collected resources about writing and research

Simulation

For networked systems research, the most popular (free) network simulator is ns-2. Tutorials can be found at here and here.

If performing operating and storage systems research and you need access to one or two computers for testing your system, please contact us. For larger-scale tests, you should try out an emulation or test-bed...

Testbeds and Emulation

Emulab

Emulab is a network emulation environment at the University of Utah. It provides racks of machines and programmable switches that can be configured to form mostly-arbitrary network topologies, with controllable delay and loss between nodes. It's a great way to test real programs in repeatable conditions or at scales that you can't get on your own.

Emulab also provides a set of wireless nodes that you can control, located around their building. They also have an experimental mobile robot testbed that could make for fun projects (the robots are designed to move computers and radios around so that you can do repeatable experiments involving mobility).

If you wish to use Emulab for your project, you should join our class emulab project.

Planetlab

PlanetLab is a large-scale, distributed collection of machines that can be used for experiments and measurements, managed by a group based here at Princeton. It has around 600 nodes scattered over about 350 sites. The machines run something linux-ish that you can login to and run programs, and there exist a variety of utilities for automatically distributing software to the nodes, running programs, and so on.

You must sign up to use a PlanetLab account. Please only sign up if you're going to use the account, since it imposes some management overhead on people not involved with the course.

Wireless

ORBIT: Large wireless testbed (400 nodes) at Rutgers. To join, see their web pages.
moteLab: 30 sensor nodes deployed in CS building at Harvard. To use moteLab, you'll need to send them an email (listed on their web page). Please CC: mfreed on the mail.
MistLab: 60 Mica2/Cricket nodes (sensor boards, much like the motes in moteLab) distributed over the 9th floor in the Stata center at MIT's CSAIL. For access, see their web page and send them an email, and please CC: mfreed on the mail.

Analysis

Books: Raj Jain, the Art of Computer Systems Performance Analysis - a very good overview of lots of mathematical techniques and queuing bits, aimed at a systems audience

Last updated: Wed Jan 23 16:21:57 EST 2008 [validate xhtml]