CS589 (Fall 2022): Advanced Computer Networks: Syllabus

Lecture: Mon, Wed 9-10:30AM in 1130 LBME
Discussion: Fri 1030-1130AM in 1130 LBME
(Zoom link, passcode: eecs589)
4 units

Instructor:


GSI:

    Ruiyang Zhu
    4617 BBB
    ryanzhu@umich.edu
    Office hour: Friday 11:30AM-12:30PM 


Announcements

Prerequisite:

CS489 or undergraduate introductory class to networking required. Background in system programming, statistics and probability are helpful. Please refer to this class from Princeton for an overview of the background networking material. See also the Schedule tab of this course from GaTech for short video overviews of many introductory networking concepts.

Overview:

This is a graduate level course on computer networking focusing on advanced topics and is a must for anyone interested in doing research in computer networks. The course consists of both a reading/lecture/discussion component and a project component. We will read around ~20 research papers on most recent topics of computer networking. This class examines the current and emerging research topics in computer networking. Topics covered include network protocols, network data center networking, measurement, Internet routing, peer to peer networks, network security, wireless and sensor networks, and more recent topics such as SDN and ML/AI as related in networking.

We focus both on the existing technologies also on why some of them are not sufficient because of technology trends or changes in fundamental assumptions. As an example, early designers of Internet assumed cooperative behavior of end nodes. The state of the art of tens of thousands of compromised hosts, i.e., zombies, completely changes this assumption and today's landscape on the Internet. Emphasis of the course is on topics in wide-area networks, wireless networks, and measurement methodologies for Internet and wireless experiments, as advanced topics in machine learning and security. Students are expected to carry out a research project including analysis, design, and implementation components when appropriate on a novel subject.

The class projects can be either of the following types: algorithm design applied to networked system, implementation of a novel networking system, measurement of existing network protocols, and simulation of a proposed network algorithm. We emphasize on problems that are real and solutions that will make a difference, ideally can be deployed. The lecture will be conducted in an interactive fashion. I will lead the discussion for the first part of the class, but I expect everyone to participate. For the second half of the class, I expect each student (can be in groups) to present a paper. You will be graded for both the paper summaries and class discussions.

Textbooks:

Please see the course's Reading List for the papers we will be reading this semester. There is no textbook for this course. For background review and reference, you may find the following textbooks useful:

Grading:

Research project 50%
Assignments/exam 20%
Paper presentation 10%
Reading summaries 10%
Class discussion 10%

zmao@umich.edu