JMU
Sample Questions for Exam 1


  1. Answer the questions from homeworks 1,2 and 3.
  2. Answer the questions from chapters 1 and 2 in the textbook.
  3. Choose the best answer to each of the following:
    (1) _____ A traditional string telephone allows for which kind of communication?
    1. Simplex
    2. Half Duplex
    3. Full Duplex
    4. Quadraplex
    (2) _____ The Domain Name System (DNS) makes use of:
    1. Authoritative Name Servers
    2. Root Name Servers
    3. Caching Name Servers
    4. All of the above
    5. None of the above
  4. Carefully define each of the following terms (as we have used them in this course):
    Client-Server Architecture


    Quotidian


  5. Identify whether each of the following is synchronous or asynchronous.
    Electronic Mail


    "Snail" Mail


    Walkie-Talkie


    Telephone


  6. Identify whether each of the following uses a channel that is simplex, half-duplex, or full-duplex.
    Electronic Mail


    "Snail" Mail


    Walkie-Talkie


    Telephone


  7. Is a CS460 lecture synchronous or asynchronous? Does it use a channel that is simplex, half-duplex, or duplex? What about a recorded lecture?
  8. Create a UML state diagram for that describes the dynamic behavior of a simple, single-threaded, non-caching HTTP 1.0 server. Your diagram must include an AcceptingConnection state, a Reading state, and a Transmitting state. It may include other states as well.
  9. Consider a challenge-response protocol in which the server sends a challenge/question to the client, the client sends a response to the server, and the server sends an acknowledgment to the client. Suppose the probability that a challenge sent by the server is not received is given by \(p\), the probability that a response sent by the client is not received is \(q\), the probability that an acknowlegment sent by the server is not received is \(r\), and that all three of these events are independent.
    1. What is the probability of a successful challenge-response-acknowledgment?
    2. Suppose \(p = q = r\). What does \(p\) (and, hence, \(q\) and \(r\)) have to be in order to ensure that the probability of a successful challenge-response-acknowledgment is at least \(0.729\)?
  10. Consider a small company that has a help desk that responds to questions using email. The help desk is staffed 24 hours per day and 7 days per week with one employee. Questions are received at a rate of 360 per day and are responded to at a rate of 480 per day. What is the expected (steady state) time in system for a question? (Note: You may make all of the same assumptions we made in lecture.)
  11. Identify two common switching modes.
  12. Explain why the Domain Name System (DNS) is considered an application layer system.
  13. Explain how non-recursive queries (i.e., referrals) are handled in the Domain Name System (DNS).
  14. Describe the steps used in a DNS poisoning attack. Why aren't they effective against a server that uses referrals rather than recursive queries?
  15. Suppose you have three computers in your apartment/house and that your ISP has allocated 5 IP addresses for your apartment/house. Suppose further that, for some inexplicable reason, rather than assign an IP address to each computer, your ISP selects an IP address at random (each of which has a probability of \(1/5\) of being selected) and assigns it to a computer each time the computer is turned on. Since your ISP does not keep track of which IP addresses it has assigned, it might assign the same IP address to more than one computer.

    Suppose two computers are turned on. What is the probability that they will have the same IP address?

    Suppose all three computers are turned on. What is the probability that any two will have the same IP address? Intuitievly, why does it make sense that your answer to this part of this question is larger than your answer to the previous part of this question?

  16. Explain why a peer-to-peer system that uses a central directory is less robust than a peer-to-peer system that uses flooding.
  17. Suppose you need to distribute a 20 Gbit file to 50 peers when the server has an upload rate of 2 Mbps, each other machine has an upload rate of 1 Mbps, and each other machine has a download rate of 1 Mbps.
    1. How much time is requried to distribute the file using a client-server system?
    2. How much time is requried to distribute the file using a peer-to-peer system?
  18. Explain the value of caching in content distribution networks.
  19. Compare central directory peer-to-peer systems with flooding peer-to-peer systems.
Department of Computer Science

Copyright 2013