Computer Networking

Test 2, 4 April 2013


There are six questions; answer at least five of them. You have one hour, from 1 to 2.

Up to the first 75 words appearing after a question are taken to be the answer to the question. Words following the 75th word are not part of the answer and will be ignored.

  1. Two systems S and R are connected by a 100 Mbyte/sec link. S, which can send 8,000 frames/sec, sends a continuous stream of 1.5 kbyte frames to R, which can accept frames at a rate of 7,000 frames/sec. S is using a sliding window to for flow control. What is the smallest window S can use to make sure the link operates at 100% capacity? (Keep in mind that "not possible" could be an answer.) Show your work.
  2. Assuming route cost is measured in hops and the objective is to have minimum route costs, how would an infinite cost (hop count) be used in a routing algorithm? Make sure you indicate the routing algorithm used in your explanation.
  3. Describe a CRC generating polynomial that provides error detection equivalent to a single parity bit. Justify your answer.
  4. True or false: any LAN in a network of bridges and LANs will be attached to at most two active bridges. Assume the network is fully operational (any host in the network can send a packet to any other host in the network) and stable (bridges and LANs are not being added to or removed from the network). Justify your answer.
  5. What would be the effect on overall throughput at the data-link level if a WiFi-like system used collision avoidance (CA) without using carrier-sense multiple access (CSMA)? Justify your answer.
  6. Assume each router in a network uses a list of packet ids to suppress duplicate broadcast packets. Describe two ways a router can manage the size of packet-id list. Assume the network is stable (no hardware is added to or removed from the network) and error free.

This page last modified on 2013 April 3.