CNT 5106 C Computer Networks: Fall 2009
Instructor: Prof. Ahmed Helmy
Note: This page is under construction and will be updated continuously.
-
Course
Description and Syllabus
- Lecture Notes/Slides:
[Note: lecture drafts maybe updated continuously to enhance the quality of
the lecture as I (the Prof.) see fit. Usually the slides already presented
in the lecture will not change with the next draft (except to make
corrections if needed, but then a note will be made)].
- Homeworks:
- Some pointers to find demo and reading material:
- The website for the required book by Jim Kurose and Keith Ross
can be accessed here, then click on
student resources.
- Particularly interesting are the applets on the above website, as you
can experiment with various parameters (of delay, distance, etc.) and
observe the network performance and behavior.
- Experiment with the following applets from the book's website at
this
link. Especially the 802.11
CSMA/CA basic applet and the 802.11
CSMA/CA with hidden terminal.
- Play the animation
for the wireless access protocols through the network animator
(nam) [see above to install/run nam].
- Videos to show some basic behavior of TCP window mechanism:
- Basic window mechanisms (e.g., selective repeat)
here.
- Slow Start (exponential increase) then congestion avoidance (linear increase)
here.
- Change of TCP (cwin, ssthresh) parameters
here.
- Video to show interaction of TCP with routing dynamics:
here.
- The
network
animator (nam) [.exe, bin for windows XP], can be used
with the following scripts:
selective
repeat (or TCP without slow start),
Self
Clocking,
TCP with
Slow Start,
TCP with
slow start (scenario 2),
TCP
with fast retransmit mechanism,
TCP
with routing dynamics.
- Light reading on elementary queuing theory: Basics,
M/M/1 model.
- Snap shots of Google Earth files obtained using Wififofum and other wireless signal
measurements around Gainesville (pic
1, pic
2).
animation of clustered behavioral
profiles in mobile networks (video)
- This
article
by the New York Times relates to the bluetooth encounter experiments you have been carrying out with the
handheld devices and cell phones. The MIT trace is available through the
Exp2 website (to be updated).
- Some resources for multicast, specifically the Protocol
Independent Multicast (PIM). Recommended initial readings: From the PIM-SM
Spec sections 1 & 2, and the Bootstrap
mechanism for dynamic Rendezvous Point (RP) election sections I, II
and IV A. Once you are done with the initial reading, if interested, you
can read the rest of the documents and also the PIM architecture paper,
before others.
- For those interested in research, a small (potentially helpful) note: How
to start research? (A personal note for networking students)
- Announcements:
- The TA office hours are as follows: (1) Q. Deng (contact.qdeng@gmail.com): Mon 3-4pm, Wed. 12:45-1:40pm, (2) Nan
Jiang (nan@cise.ufl.edu): Tuesday 12:35-2:35pm, (3)
Ruijin Wu (cszwrj@gmail.com): Thursday 3-5pm, (4) Gautam Thakur (gsthakur@cise.ufl.edu): Mon and Wed. 4-5pm. The TAs will
conduct their
office hours
in room E309 (third floor
CISE Bldg).
- The TAs should be available to answer questions in-person or by email during office hours. Gautam Thakur will be the main
contact regarding the experiments, checking out devices and the wireless mobile traces.
- Homework 1 is now on-line, due on Oct 7th.
- The midterm exam is going to be held on October 21st, at TUR L005, TUR L011 from 8:20pm to 10:10pm. These
were the only available rooms (and times) for such a big class. More info to follow soon ...
- Homework 2 is now on-line, due on Oct 19th. [Note: for students who didn't get to see the ATM part - the Edge tape
ran out right at that point - you do not have to submit Q11 on Monday, you have until Wed. to submit it. However, I
would recommend you try it out, it doesn't rely on ATM-specific knowledge and you should be able to do it.]
- Here are the last 2 pages of the midterm containing formulas you
may need. This will be included in your midterm paper and no need to bring it. Familiarize yourself with it in case you
need to use it in the exam.
- Office hours for Oct 19 (Monday) will be delayed until 2:30pm since I have to chair a town-hall meeting for the
engineering faculty from 12-2pm.
- How do I prepare for the midterm? My recommendation is to go over the class notes (slides and your own notes during
discussions) first, then read (as needed) from the book those parts for which you may want clarification or more
details. The homeworks are also important. Do you have to read all the chapters covered in the book? No. The parts of
the book not covered in the lecture
are not required. But there are some parts covered in the lectures (e.g., queuing theory) that you will not find in the
book in detail. The exam is closed book and closed notes, but I will include 2 pages of formulas at the end of the exam
(please find those 2 pages Here). Focus more on understanding, rather than
memorization.
- Based on the feedback I got from students, the format will be a mix of short-answer questions, calculations and
design questions. The questions may come in no particular order of difficulty, so start with those you know. I plan to
have multiple models of the exam, so the person next to you may have different questions (or different answers, or
different ordering or parameters) than you. Integrity is demanded of everyone and no cheating will be tolerated. Think
carefully before you answer and write only as indicated by the space given in the exam paper, not more. Let
your efforts show.
Good luck to you all!
- Intuitive explanation of the queuing question Q4 in hwk2: we know (from the lecture) that the average queue size is the
same
for both systems (the original system with 1 server, and each of the systems with parallel servers) if the utilization is the
same (for M/M/1 and M/D/1). Since the system with 1 server has twice the capacity, then the delay experienced by elements (or
packets) in that system is half as the delay experienced by elements in each of the other parallel systems.
- Expriment assignment created on the LSS. Now, student groups can view the experiment document and submit the report. Thank
you Gautam !
- Next week I am on travel and will not be able to give the lecture myself. Instead we may have a 'workshop'-like
presentation for the experiments and projects by Gautam and other PhD students of mine. Details to be announced on LSS (the
e-learning system). Please contact Gautam for more details. Thanks Gautam (again) and the rest of the Nomads group!
- There will be class as usual on Mon Nov. 9.
- All student groups who are doing the measurements experiment hould send their measurements and traces to Gautam (gsthakur@cise.ufl.edu), thanks!
- [Nov 18] A new homework is posted for the Data link layer and wireless networking (Homeworks 3 and 4).
Note that they are combined in one homework that has the weight of two homeworks. It also has extra points
for some of the questions that may have an open ended answer. It is due on Dec 2nd to the TAs.
- The final exam date is planned for the same date posted by the registrar office: Thursday Dec 17, 09,
12:30-2:30pm. Room(s) to be determined at a later date.
Office Hours: office hours are Mon. 11:30-12:30, Wed. 9:30-10:30 in CSE426
[Office hours are updated every week. If you can't come to the above
office hours send me email and I'll setup an appointment at another time
for you.]
