a) Name the five layers of the Internet protocol stack and give an example of protocol for each layer (except for the lowest layer). Do not use abbreviations or acronyms.
b) Switch tables can get large for large LANs since each station requires one entry. In order to reduce the number of entries, your colleague suggests using prefix aggregation for Ethernet MAC addresses similar to how it is used in CIDR for IP. What do you think of this suggestion?
c) What are the three most common packet sizes in the Internet? Explain why this is the case.
Question 2
Answer the following questions regarding transport protocols and their performance. Assume a network uses 100 Mbps links with a 1 ms one-way end-toend delay between two nodes. Also assume that these nodes use a sliding window transport protocol with a maximum window size of 3 packets and a fixed packet size of 1300 bytes (50 bytes of headers and 1250 bytes of payload). Assume acknowledgement packets are 50 bytes in size. There is no connection setup.
a) What is the throughput of this protocol from the point of view of the endsystem application (i.e., how much payload data can the protocol transfer per second)? Assume queuing and processing delays are zero, but consider transmission delays. Please report your results in bits per second, not bytes per second.
b) What changes could be made to the transport protocol to increase the throughput?
c) What window size would be necessary to achieve 80 Mbps of throughput from the point of view of the end-system application?
Question 3
Answer the following questions regarding IP address allocation and routing tables.
a) Consider the network topology shown below. Assume you are given an address block of 128.119.86.128/25. Assign a suitable address to each network interface (enter IP address in white rectangular box). Allocate addresses such that they can be aggregated into prefixes. Note the network prefix in the while oval