COMP/ELEC 529

COMP/ELEC 529 - Computer Network Protocols and Systems
Rice University

Fall 2009

Instructor

Prof. T. S. Eugene Ng
Office: Duncan Hall 3005
Email: eugeneng at cs.rice.edu

Meeting

1:00pm - 2:20pm, Tuesdays & Thursdays
Room: Brown College 145

URL

http://www.cs.rice.edu/~eugeneng/teaching/f09/comp529/529.html

Course Overview

Computer networking is a rapidly evolving discipline with many exciting opportunities. Challenges that arise in networking and particularly in the Internet tend to be truly global problems that impact millions of users. However, existing solutions to many of the classical challenges (e.g. routing, congestion control, quality-of-service, management) remain unsatisfactory, while new challenges (e.g. worm, denial-of-service, peer-to-peer) are emerging as people continue to use the Internet in ways the original designers did not anticipate in the 70's and 80's. In the long run, coping with the complexity of the Internet may become the ultimate challenge.

The goals of this graduate level course are to provide students with a proper grounding in the basic concepts and seminal work in computer network protocols and systems, and to introduce students to some of the most exciting recent research developments. The basic concepts covered include network architecture, switching, routing, congestion control, and quality-of-service. The advanced research topics will focus on network structural measurement and modeling, network scaling and management, overlay networking and peer-to-peer systems, network security, and new network architectures.

Course Format

The course will consist of lectures and discussions, paper readings and reviews, and a group research project. The lectures will begin with a broad overview of the basic concepts in computer networking and gradually transition to the advanced topics. Each lecture will be based on one or two research papers. Students must read the assigned papers and submit paper reviews before each lecture. Students will also form teams of two or more, each team will tackle a well-defined research project during the semester. A list of suggested project topics will be provided. All projects are subjected to approval by the instructor. The project component will include a short written project proposal, a short mid-term project presentation, a final project presentation and a final project report.

Prerequisites

Undergraduate experience with operating systems, computer networking, and algorithms is required.

Grading

Paper reviews	40%
Project proposal, presentations, and final report	40% (Project grading breakdown: proposal 5%, mid-term presentation 20%, final presentation 25%, final report 50%)
Class participation	20%

Writing and Submitting Review

All students must read the assigned papers and write a review for one of the assigned papers before each lecture. Email the review to the instructor (eugeneng at cs.rice.edu) prior to each lecture. The instructor will provide feedback and grades to students.

Do not send reviews as email attachments. Please send one review in plain text per email in the body of the email message. In your email, please use the following format for the subject line to enable some automation:

comp529review:[PaperID]

Recognizing that not everyone has had experience writing reviews, the reviews on or before September 29 will be graded on a credit/no-credit basis.

A review should summarize the paper sufficiently to demonstrate your understanding, should point out the paper's contributions, strengths as well as weaknesses. Think in terms of what makes good research? What qualities make a good paper? What are the potential future impacts of the work? Note that there is no right or wrong answer to these questions. A review's quality will mainly depend on its thoughtfulness. Restating the abstract/conclusion of the paper will not earn a top grade. In reading papers and writing your reviews, you may consider the following questions:

What are the 3 most important things the paper says? These could be some combination of their motivations, observations, interesting parts of the design, or clever parts of their implementation.
- from author's perspective at the time
- from your perspective with the benefit of hindsight
Describe the paper's single most glaring deficiency. Every paper has some fault. Perhaps an experiment was poorly designed or the main idea had a narrow scope or applicability. Being able to assess weaknesses as well as strengths is an important skill for this course and beyond.
What were the key observations that had led to the research direction? Where did the observations come from? Personal experience? Work environment?
What were the key obstacles or risks that could have presented the research from being successful? Were the researchers aware of them? How were these obstacles eventually overcome?

When reading the papers, it is important to understand the contexts in which the research was done. First, check out the year of the paper. Then you may want to consider the following contexts:

Intellectual context: leaders in field, major research directions/themes at the time, assumptions that are taken for granted at the time. Best way is to glance through the relevant ACM conferences in the previous couple of years.
Industry context: competitive landscape, technology/industry trends, primary industry concerns.
Personal context: researcher's stage of career, researcher's physical and people environment

Project

The course project is intended to be a mini-research project that allows each student team (2 or more people) to explore an aspect of computer networking that is of interest during the course of the semester. There will be four milestones in the project component: (1) project selection and proposal, (2) mid-term presentation, (3) final presentation and, (4) final report. Further details will be provided during class.

Reference Textbook

Computer Networks: A Systems Approach, 4th Edition
by Larry Peterson, Bruce S. Davie
Publisher: Morgan Kaufmann; 4 edition (March 8, 2007)
ISBN: 0123705487

ADA

If you have a documented disability that will impact your work in this class, please contact me to discuss your needs. Additionally, you will need to register with the Disability Support Services Office in the Ley Student Center.

Schedule

Under construction, subject to changes. The reading list with links to papers is here.

Class	Date	Topic	Primary paper(s) (choose at least one paper per class to write a review)	Other paper(s) (no review needed)	Question of the day	Relevant COMP 429 Notes	Remarks
1	8/25	Overview
2	8/27	TCP/IP	[CK74]	P&D 1	Cerf and Kahn deserve winning the Turing Award
3	9/1	Internet Design Philosophy	[Cla88],[SRC84]		The end-to-end argument is obselete
4	9/3	Intra-domain routing	[FT00],[QYZ+03]	OSPF Tutorial,[AJY00],P&D 4.2.1 - 4.2.4	Overlay routing is dangerous	lecture10.ppt
5	9/8						No class
6	9/10	Inter-domain routing	[S+99],[LAB+00]	[LMJ97],[GSW02],[Pax96],P&D 4.3.3 - 4.3.4	Inter-domain routing is not essential	lecture11.ppt
7	9/15	Rethink network control	[CCF+05],[YMN+07]		Centralized network control is a bad idea
8	9/17	Ethernet LAN Robustness	[ECN07],[ECN09]		Ethernet: Can't live with it, can't live without it	lecture8.ppt
9	9/22	Rethink Ethernet	[KCR08],[CFP+07]		Ethernet is dead
10	9/24	Data center	[CZR+09],[GCL+09]		Data center: Hot or Not?
11	9/29	Congestion control I	[Jac88],[CJ89]	[RCJ88],[Jac90]	TCP is good enough	lecture15.ppt
12	10/1	Congestion control II	[FJ93],[LAJ+03]		RED is good for the Internet	lecture16.ppt	Letter grading for reviews begins
13	10/6	Rethink congestion control	[KHR02]		XCP is vulnerable to attacks
14	10/8	Quality of service I	[DKS89],[PG93]		The Internet needs ubiquitous fair queueing	lecture18.ppt	Project proposals due
15	10/13	Midterm Recess					Midterm Recess
16	10/15	Quality of service II	[SZN97]		H-FSC is overkill
17	10/20	Rethink quality of service	[SSZ98],[SZ99]		WFQ is better than core-stateless
18	10/22						No class
19	10/27	Network security I	[SWK+00],[PWS+07]		Is denial-of-service a network problem?
20	10/29	Network security II	[MSV+03],[CCC+05]		Is worm a network problem?
21	11/3	Mid-term project presentations					Mid-term project presentations
22	11/5	Overlays I	[SMK+01],[CRS+02]		Are overlays the solution to the Internet's ossification problem?
23	11/10	Overlays II	[SAZ+02],[CDK+03]		Are DHTs useful in the real world?
24	11/12	Network Structure I	[FFF99],[LBC+03]		Does the Internet really follows power-law?
25	11/17	Network Structure II	[LAW+04],[NZ02]		Are there useful models for the complex Internet?
26	11/19	Active Network	[Wet99]		Active network: Brilliance or insanity?
27	11/24	Net Neutrality	[Cro07],[ZMZ08]		Net neutrality is good?
28	11/26	Thanksgiving Recess					Thanksgiving Recess
29	12/1	Final project presentations					Final project presentations
30	12/3	Going Forward	[Cla00],[CWS+02]		Will the Internet undergo dramatic changes?
	12/4	Final project reports due					Final project reports due