HINF 6220 - Assignment 2
Note: You may have to do some research to answer these questions. All your answers should be self-contained and aimed at a non-technical audience, for example a non-technical health professional, that has no more than the basic experience of using email, word processing and browsing and searching the Web. Make sure your answers are free of acronyms, and technical jargon, which should be defined in intuitive terms when it is used for the first time. Do not exceed the indicated page limit, measured based on an 8.5x11 page, single spaced, with 1in margins, 12pt size in Times-new-roman font. Use Web search to find the information you require. Wikipedia may be useful. Remember to use your own words in all your answers. Paraphrase, do not copy from Web sources. Reference all your sources. The page limit does not include references.
Submission instructions:
Please submit your assignment electronically by emailing it to the instructor.
Due time is 12 midnight on the due date posted. If no due date is posted, the
date when the next assignment is posted is the due date of the previous assignment.
Use the following format for the subject line:
Subject: HINF6220 - Assignment N (where
N=1,2,3,4)
Your submission should consist of one file in pdf format.
Use Acrobat (the full version, not the Reader), if you already have it, or one
of the free converters (1 2).
About the Portable Document Format
( pdf)
The name of the file should be the following:
HINF6220-AN-loginname.pdf , where
loginname is your cs login name,
and the N in AN is 1,2,3,4
Grading:
Each question will be graded with a letter grade, based on quality of content,
quality of writing and demonstration of understanding of key concepts discussed
in the lectures. Your answer should be self-contained and easy to read. The
overall grade for the assignment will be a weighted average of the individual
grades, where the weight is proportional to the size limit of each question
(not counting program outputs, if any). Letter grades are being averaged using
their GPA equivalent. No rounding takes place.
The meaning of the grades is as per the University
Calendar, section 17.1.
This style of grading will be used in all evaluation components of this course.
1. File transfer. Assume an ftp transfer of a 10MB file from a host in Halifax to a host in Beijing. b = bit, B = byte, 1B = 8b.
a. What is the propagation time? The geodesic distance between the two cities
(i.e. the distance measured on the great circle that contains both cities) is
10600km.
Use the speed of light in optical fibre (the figure is avaiilable on Wikipedia).
b. What is the time to transmit, for each of the link types on slide 19 of the lecture slides?
c. If the "pipe were full", how much data in transit would be held in it?
2. Transmission of audio data.
a. The maximum acceptable delay in voice communication is 200ms. Compare this figure with the minimum delay of transmitting 200ms of speech on a voice-over-ip (VoIP) connection that can be achieved over a line with bandwidth 2Mbps between Halifax and Beijing. Assume a rate of 32Kbps for voice. Calculate the transmission and propagation delay (assume negligible processing and queuing delay). (max. 1 page)
b. The lecture is recorded digitally. The lecture lasts 75min, and the data rate of the recording after compression is 32Kbps. The sampling rate is 32KHz. Each sample requires 8 bits to store if no compression is used. Explain the observed size of the recorded file of a lecture in bytes. What would be the size if there was no compression? (max. 1 page)
3. Growth patterns in computing. Review the specifications
of the TRS-80 computer and accessories,
and compare its cost/performance ratio with what you can buy today.
How close does this comparison come to the growth figures shown in the lecture?
You may want to use the cost-of-living
calculator for more accurate comparison of the prices between the early
'80s and 2007. (max. 1 page)
4. IP addresses. IP version 6 allocates 128 bits to an IP address, compared with 32 bits of IP version 4. Assuming classless routing in both cases, what is the maximum number of hosts that can be addressed in IPv6 and IPv4? Explain why 32 bits is too small a number of bits to allocate to an IP address. Assume that the land surface of the earth is divided into a number of equal-sized cells that is equal to the total number of addresses supported by the protocol. Each cell would then be given its own IP address (assuming that it contained a host connected to the Internet). What is the area of a single cell in IPv6 and IPv4 addressing? Assume that the radius of the earth is 6360km, the surface of the earth is equal to four times the area of a great circle, and the land surface area is 29.2% of the total surface area of the earth. (max. 1 page)
5. User Datagram Protocol (UDP). UDP is a lightweight transport protocol that offers fewer guarantees than TCP. It also lacks a congestion control mechanism. Research UDP on the Web (Wikipedia has an article that can serve as the starting point), and explain in a layperson's terms how UDP differs from TCP and how congestion can be avoided if a large amount of UDP traffic is injected into the network.(max. 1 page)