Network Measurements, Modeling and Simulations

1
Network Measurements, Modeling and Simulations

• Kun-chan Lan
• CSIE NCKU
• http//www.csie.ncku.edu.tw/klan
• klan_at_csie.ncku.edu.tw

2
Vital Information

• Course Network Measurements, Modeling
  and Simulations
• Taught by ???
• Credit 3 units

3
A little background about me

• New faulty at CSIE
• BS, NCKU, 1990
• MS, State Univ of New York (SUNY) at Stony Brook,
  1997
• PhD, University of Southern California (USC),
  2004
• Researcher, National ICT Australia, 2007

4
My research

• Mainly in Computer Networking
• My past research
• Internet traffic modeling
• Network simulation
• My current research
• Vehicular network
• Wireless network

5
For Most of You

• The first non-English course taught in English
• All communication in English, including
• Lectures
• Project presentation and report
• In/off-class interaction

6
Be Aware

• Project proposal/report are accepted only in
  English
• Credits are granted only when the English is
  comprehensible

7
Though to avoid confusion

• Please make sure your names are clearly written
  in Chinese.

8
Why in English?

• Easier for me to prepare and communicate
• To avoid awkward translation of technical terms
• You get first-hand information you get an
  internationally-consistent view of the technical
  phrases.
• and no matter what you do in the future, its
  good for you to be fluent in English anyway

9
Look at the bright side

• You get to learn things in an English-based
  environment without going abroad physically

10
But to make it easier for you..

• You are allowed to ask questions in Mandarin

11
Some Suggestions

• Read the corresponding material/chapters before a
  lecture
• Review the recording of a lecture speech
• Lectures will be recorded and put available in
  mp3 format

12
Further Alternatives

• The deadline for dropping a course this semester
  is April 23.
• This course will be offered every year

13
How many of you still wants to sign up for the
course?
14
Today

• All about the class admin

15
Roadmap

• The essentials
• Administrative Information
• Content
• Course objective and scope
• Syllabus
• Your responsibility
• Paper review
• Projects
• Grading policy
• Class material

16
The Essentials

• Course page
• http//www.csie.ncku.edu.tw/klan/ns/ns.html
• It is your responsibility to frequently check the
  Announcement link
• Kun-chan Lan
• http//www.csie.ncku.edu.tw/klan
• Click the Teaching link
• Then, click the Network Measurements, Modeling
  and Simulations link

17
Roadmap

• The essentials
• Administrative Information
• Content
• Course objective and scope
• Syllabus
• Your responsibility
• Homework
• Projects
• Grading policy
• Class material

18
Lecture Info

• Location
• CSIE Room 4261
• Time
• 910 1200
• 2 breaks
• 10-1010
• 11-1110
• Please note that during the breaks the priority
  goes to the calls of nature.
• Questions will be addressed afterwards.

19
The Instructor

• Kun-chan Lan
• Office CSIE, Room 4237a
• Inside Laboratory for Experimental Network
  System (LENS)
• Phone [phone] x62550
• Email klan_at_csie.ncku.edu.tw
• Homepage http//www.csie.ncku.edu.tw/klan

20
Office Hour

• 1-2 pm Friday
• 3-4 pm Tuesday
• Or, by appointment

21
The TA

• Yet to be determined

22
Background Knowledge

• Fundamental courses on data networks and know
  basic things from layer 3 and above
• There will be networking terms that I assume you
  already knew
• Youll definitely need to know how the TCP/IP
  network works and some research-oriented ideas.
• This will be a difficult course if you do not
  have sufficient background.

23
Recommended reading

• If you dont have sufficient background, try to
  review the following books
• Larry L. Peterson Bruce S. DavieComputer
  Networks A Systems Approach, 3rd Edition, Morgan
  Kaufmann, ISBN 1-55860-833-8
• Computer Networking A Top-Down Approach
  Featuring the Internet, 3rd EditionJames F.
  Kurose Keith W. Ross, Addison Wesley, ISBN
  0-201-97699-4

24
Software Skill

• Unix!
• C/C
• Experience with scripting language

25
Equipment Requirement

• A laptop/PC of your manipulation
• Installing Linux/Freebsd
• Installing and trying the tools
• Doing the homework and the project

26
Roadmap

• The essentials
• Administrative Information
• Content
• Course objective and scope
• Syllabus
• Your responsibility
• Assignment
• Project
• Grading policy
• Class material

27
Nature

• An advanced course on how to do networking
  research
• The typical research cycle
• Tools for networking research
• For graduates students who are
• Serious
• Responsible
• Spontaneous

28
Objectives

• Targeting audience graduate students who are
  interested in pursuing research in networking
• Aim introduce the whole process of an
  experimental research cycle from problem
  identification, proposing solution, experiment
  design, to finally evaluate the proposed
  solutions in the simulations or emulation.
• We will cover basic techniques in experimental
  design, measurements, modelling and simulations.
  A number of tools frequently used in network
  research will be introduced.

29
Scope

• Measurement strategies and techniques
• Principles of network modelling
• Fundamental of experimental design and
  performance evaluation
• Publicly available test-beds Orbit, Planetlab,
  Emulab
• A collection of network research tools
• Measurement tool ping, traceroute, iperf
• Collection tool tcpdump, ethereal
• Traffic modelling SURGE, RAMP, Swim
• Topology modelling Rocketfuel, GT-ITM
• Path characteristics estimation IGI and
  pathChirp
• Simulation NS-2
• Emulation Nisnet, dummynet

30
The aim of this course

• At the end of semester, I hope that you will
• Familiarize with the typical research cycle
• Know what are the available tools you can use for
  your research
• Know how to write and present your work in English

31
Syllabus Part I
week 1 (2/26) Administration issues -- group list
due week 2 (3/5) measurement strategy and tools
week 3 (3/12) no class (Instructor on-leave),
paper review list due week 4 (3/19) testbed
week 5 (3/26) simulation ns-2 (I), project
proposal due week 6 (4/9) in-class project
proposal presentation week 7 (4/16) simulation
ns-2 (II), Quiz 1 week 8 (4/23) simulation ns-2
(III)
32
Syllabus Part II
week 9 (4/30) emulations, Quiz 2 week 10 (5/7)
midterm review week 11 (5/14) traffic modeling
week 12 (5/21) topology modeling week 13 (5/28)
wireless modeling week 14 (6/4) bandwidth
estimation week 15 (6/11) Misc week 16 (6/25)
In-class final project presentation
33
Roadmap

• The essentials
• Administrative Information
• Content
• Course objective and scope
• Syllabus
• Your responsibility
• Assignments
• Term project
• Grading policy
• Class material

34
Deliverables

• Paper review and presentation (2) 20
• In-class quiz (2) 10
• Group Project 70
• No written exam in this class

35
Homework

• 4 assignments
• Reading-based (20)
• 2 paper reviews from the reading list on the
  course web page
• In-class presentation
• Tool-based (10)
• in-class quiz
• show you know how to use the tools

36
Paper review list

• Pick 2 papers that interest you from the reading
  list
• If none of the papers in the reading list excit
  you, find your own (but get my consent)
• The paper you review should be related to the
  project youre going to do
• Send me the list by the end of week 3 (3/12)

37
In-class quiz

• Simulator
• Emulator
• Measurement and data collection tools

38
How to review a paper

• What are the major issues addressed in the paper?
  Are these issues important?
• Novelty and creativity of the paper?
• Technical depth of the paper?
• What are the strengths of the paper?
• What are the weakness of the paper? Can you
  improve the paper?
• Writing style and readability?

39
Term Project

• A research-oriented exercise
• Research cycle
• Literature review
• Problem definition
• Comparison of existing solutions
• Potential ways to improve
• Experimental design
• Analysis
• Presentation

Phase I
Phase II
40
Phase I

• Skim through papers provided in the reading list.
  
• Select a topic of interest.
• Read assigned papers in that topic
• Find a problem that needs to be solved (any
  problem)
• Compare existing solutions (additional papers
  welcome)
• Suggest solutions (can be existing ones or
  propose one yourself)

41
Phase II

• Identify an evaluation problem
• Hypothesize results
• Propose an experimental plan
• Execute the plan
• Analyze experimental results
• Present the results

42
Formality

• Working in groups 1 or 2 or 3
• Write-up
• 1-page single-spaced proposal at the mid point of
  the semester
• 6-page, 10-pt-font, single-spaced report by the
  end of final week

43
In-person Communication

• Each group should arrange a time to meet with me
  after turning in the proposal
• I will give you my feedback on the proposal

44
Proposal Presentation

• Each group will present your project idea
• The class and I will give you our feedback on the
  idea
• Each group will refine and submit the project
  proposal

45
Project Presentation

• Each group will submit a project report
• Each group will present your project work
• The class and I will give you our feedback on the
  work
• Presentations are peer graded

46
Two types of research project

• Survey paper (phase I)
• Compare all the existing solutions to one
  particular problem
• Identify their pros and cons, and propose your
  own solutions
• Extra credits are given if you can show your
  solutions perform better than any other existing
  solutions (phase II)
• experimental paper (phase II)
• propose a solution or a hypothesis for a given
  problem
• carry out some experiments to test your solution
  or hypothesis

47
Picking the project

• It is strongly encouraged that you select a topic
  of your own research interest!
• You will not come out with something cool if you
  are not passionate about it
• But if you are just clueless

48
Suggested project topics

• Survey and comparison of existing network coding
  technique
• Survey and comparison of existing opportunistic
  networking technique
• Survey and comparison of existing VANET data
  dissemination protocols
• Survey and comparison of existing time
  synchronization for multi-hop
• Survey and comparison of existing bandwidth
  estimation techniques
• Survey and comparison of existing load balancing
  techniques for wireless mesh network
• Smart parking parking space availability
  dissemination on multi-hop wireless network
• Road congestion detection using probing car and
  vehicular ad-hoc network
• Gateway selection in wireless mesh networks
• On the reliability of wireless mesh network
• Comparison of VoIP performance over Skype and
  MSN
• On the origin of traffic burstiness
• Measure and model wireless links on a street
  testbed
• The impact of codec on the quality of VoIP

49
Roadmap

• The essentials
• Administrative Information
• Content
• Course objective and scope
• Syllabus
• Your responsibility
• Assignment
• Term project
• Grading policy
• Class material

50
Grading

• Paper review and presentation (20)
• In-class quiz (10)
• Project proposal presentation (10)
• Mid-term review (10)
• Project proposal (10)
• Final project presentation (15)
• Project report (25)
• Extra credit
• Class participation
• Up to 10

51
Evaluation of Quiz

• Correctness
• For example, I might ask you to simulate a
  20-node network and output your trace to a file.
  And then I might compare your trace with the
  expected output.

52
Evaluation of paper review

• Do you describe the motivation of the paper?
• Do you describe the methodology used in the
  paper?
• Do you point out the strength of the paper?
• Do you point out the weakness or limitations of
  the paper?
• Is your presentation understandable?

53
Guideline for writing project proposal (I)

• Motivation
• Why the problem you're trying to solve is
  important? Who will care?
• What are the research issues? Why the problem
  still remains unsolved?
• Related Work
• Have you done a literature survey of all the
  previous related work (hint Google is your
  friend!!)?
• What's NEW in your project?
• Methodology
• Description of Architecture/Protocol/Algorithm
  you plan to implement?
• The novelty and limitations of your methodology?
• What are your assumptions?

54
Guideline for writing project proposal (II)

• Description of your experiments
• How do you model the network (e.g. traffic,
  topology, links)
• What are your model parameters?
• How do you collect the data? (e.g. describe your
  sampling methods if there is any)
• What are your performance metrics?
• Expected outcome
• What are the expected results from your
  experiment?
• What are the possible obstacles? What is your
  plan B (i.e. your alternative solution)?
• What are the tasks you need to accomplish to
  meet your project objectives?
• What is your project timeline?

55
To be fair for the grading

• For group project, the responsibility of each
  group member should be stated clearly in the
  project proposal

56
Guideline for presentation (I)

• Before preparing any slides, identify 2 to 3 key
  things you want the audience will remember for at
  least 24 hours
• If your average time per slide is X minute, and
  your allotted time is Y minute, limit the number
  of slides to Y/X

57
Guideline for presentation (II)

• TO-DO
• Make sure you tell the audience the following
  about your work
• motivation
• approach
• key results
• Make sure you label your x-axis and y-axis on
  every plot
• Use graphics or animation if you can
• One picture says thousands of words
• NOT-TO-DO
• Bore the audience with too much details
• Use very small fonts
• Run the talk over-time

58
Group project evaluation criteria

• Is your literature review comprehensive?
• Creativity
• Functionality
• Novelty (or Originality)
• Has the outcome of your project any impact to
  other researchers?
• Peer rating for the presentation
• Audience will participate the rating of the
  project/proposal presentation
• Each student in the audience will write down the
  pros/cons of each presentation and submit
  electronically to TA
• Documentation
• How readable are your source codes? Are they
  modularized and well documented?
• The readability of project proposal and final
  project report.

59
Mid-term review

• Is your progress satisfied? Is your project still
  on track?
• How did you handle unexpected obstacles?
• Is your project still going toward the same
  direction? Or you have changed to a new
  direction?

60
Academic Integrity

• Respect yourselves
• Any cheating behavior will be granted ZERO

61
Roadmap

• The essentials
• Administrative information
• Content
• Course objective and scope
• Syllabus
• Your responsibility
• Homework
• Term project
• Grading policy
• Class material

62
Class Material

• No textbook
• Tool manual pages
• Reading List

63
Additional Material

• Lecture slides in ppt format
• Lecture recordings in mp3 format
• All available from the course page

64
Questions?
65
Select your group members now!

• Find your group members after the class
• Email your group members (name, ID) to me before
  COB (Close Of the Business day) of today!
• For those could not find partners, I will
  randomly assign you to team up with somebody