Advanced Network Management-spring2015

Announcements

Jun. 10: Alternative homework due date is 5pm June 25th.
Jun. 10: Demo & presentation due date is June 17th.
Jun. 10: Technical report due date is 5pm June 15th.
Mar. 10: Project proposal due date is postponed to April 15th.
Mar. 2: Each student please send your email address to [email protected], with subject line “ANM student”.
Mar. 2: Website is online!

General Information

Course

Class Time

9:50am-12:15pm Wednesday

Room

Six-Jiao 6B-311

Important Dates

Project Proposal Due: April 15th.
Technical Report Due: 5pm June 15th.
Demo & Presentation: June 17th.

Personnel

Instructor: Dan Pei

Associate Professor

Department of Computer Science and Technology

Tsinghua University, Beijing, China 10084

+86(10)62792837

Office: East Main Building 9-319

Mail: [email protected]

Teaching Assistant: Ping Liu

Ph.D. Student

Department of Computer Science and Technology

Tsinghua University, Beijing, China 10084

Mail: [email protected]

Course Description

Course Number

80240663

Credit

Required text

None.

Reference texts

Computer Networking: Top-Down Approach (6th Edition)
by James F. Kurose, Keith W. Ross.
Statistical Data Mining Tutorials
by Andrew Moore.
Learning From Data
by Yaser S. Abu-Mostafa, Malik Magdon-Ismail, Hsuan-Tien Lin.
Time Series Analysis: Forecasting and Control (4th Edition)
by George E. P. Box, Gwilym M. Jenkins, Gregory C. Reinsel.

Prerequisites

Familiar with at least one programming language. Undergraduate Computer Networks Course.

Grade

Your grade is based on your following performance:
Technical Report: 20%. Due 5pm June 15.
Demo & Presentation: 50%. During the Lecture on June 17.
Homework: 30%. Due 9:50am Wednesday before each of six lectures you picked.

Homework

Our homework format is meant to help the students read the research papers with critical eyes. Each student is required to write critiques for 6 papers in total throughout semester. For each of the twelve lectures (weeks 2,3,5,6,7,8,10,11,12,13,14,15), the instructor will post a few papers on the website by 5pm on the Friday before the lecture. Each student can pick by yourself AT MOST ONE paper out of the papers for each lecture. For each picked paper, based your own understanding, you need to write (in your own words) three contributions of the paper, and three limitations of the paper, and send the critique to our TA via email BEFORE the lecture time (9:50am) of your picked paper. Late submission does not count as submissions

Alternative Homework

Reproduce 3 figures in the papers we covered in the lecture based on the data shared by the authors.
You have to submit the figures the intermediate results in each step you took. Each such homework counts as three critiques.
Below are a few candidate papers:
1. 2014NSDI-How speedy is SPDY?
2. 2013IMC-Measuring and Mitigating Web Performance Bottlenecks in Broadband Access Networks
3. 2013IMC-Peeking behind the NAT
4. 2002Mobicom-Analysis of a Campus-wide Wireless Network
5. 2006INFOCOM-Extracting a mobility model from real user traces_slides
6. 2007SIGCOMM-Automating Cross-Layer Diagnosis of Enterprise Wireless Networks
7. 2013Mobicom_Observing Home Wireless Experience through WiFi APs.
8. CoNEXT2010_MAWILab: Combining Diverse Anomaly Detectors for Automated Anomaly Labeling and Performance Benchmarking.

Project

Students form project team one or two students. Project proposal is due on April 15th during the lecture, during which each team has 5 minutes to give a talk (with slides) to talk about team members, project idea, the basic approach, expected results, and schedule. A team can pick a topic from the below list of project ideas, or propose a topic by yourself. In either case, the team needs to talk to the instructor and/or the TA about your draft proposal BEFORE the April 15th class presentation. Each team is required to submit a project technical report (PDF format) by 5pm on June 15th. On June 17th during our lecture time, each team is expected to give another (and longer) talk about your finished project, and give a live demo of your project during the talk.

Syllabus

This course is a graduate course and is primarily project-oriented. It aims to teach students how to build REAL systems that measure the REAL data from the networks and services, process the using Big Data techniques such as Machine Learning, and solve their REAL performance and security problems.

Through case studies based on recent research papers in top network conferences, this course will cover the latest research progress in network management in these areas: measurement, anomaly detection, diagnosis, and mitigation. Along the way, we will also briefly review techniques that have broader applications more than just network management, such as time series analysis, association rule mining, and machine learning.

This course focuses on how to improve the performance of Mobile Internet:

Targeted Services: Web-based Services such as search engine, online shopping and social networking; Video Streaming Services.
Targeted Networks: Enterprise WiFi Network, Residential WiFi & Broadband Networks, Cellular Networks, and Data Center Networks.
Targeted Devices: Smart Phones.

Lectures

Week	Date	Topic	Slides&Reading lists
1	Mar 4	Course Introduction & Network Basics	week01
2	Mar 11	Web-based services	week02
3	Mar 18	Web-based services	week03
4	Mar 25	No class
5	Apr 1	Web-based services	week05
6	Apr 8	Video streaming	week06
7	Apr 15	Video streaming (Project proposal Due)	week07
8	Apr 22	WiFi	week08
9	Apr 29	No class
10	May 6	Residential WiFi	week10
11	May 13	EWLAN	week11
12	May 20	Smart phone 1	week12
13	May 27	Smart phone 2	week13
14	Jun 3	Anomaly detection	week14
15	Jun 10	Troubleshooting	week15
16	Jun 17	Project Presentation (Project Report Due: Jun 15th 5pm)

Tsinghua NetMan Lab