Q1. Answer, why or why not, the following? (Any 12)                    (12)

1.Class D addresses are kept for multicasting

2.Addresses specify network connections and not an individual computer

3.Directed broadcast is same as limited broadcast

4.Routing is same for the same host even when used with different IP addresses

5.The addresses above 247.255.255.255 are reserved

6.Address resolution through dynamic binding is far better than through direct mapping

7.RARP is confined to a single LAN but ARP can be used with multiple LANs

8.Identification field is very important for a fragmented packet

9.The data carried by an IP datagram has no effect on the header of IP datagram

10.Direct delivery is always included in any transmission

11.Next hop routing is not efficient

12. ICMP is error correcting protocol/mechanism embedded in the network layer of TCP/IP protocol suite

13.Using ICMP, all kinds of errors occurring in either physical, data link or network layer can be trapped 

14.Source quench message is sent to request sender to stop transmission in case of congestion

15.Parameter problem is related to corrupted header

Q2. Answer the following (Any 8)                                                       (8)

1.What is the idea of copy bit in the code field option of the datagram?

2.Time to live is a field does not bear the same meaning.

3.How the receiver, in case of fragments, checks length of the datagram?

4.How ARP is refined in such a way that not only receiver, others are also benefited?

5.What is the purpose of Network Byte Order?

6.What is the significance of address 127.x.y.z?

7.Why IP addresses are kept in the routing table instead of physical addresses?

8.Why hosts should not route the datagram?

9.What is the importance of redirect ICMP message?

10.What is the difference between Destination Network unreachable and unknown message? When can each occur?

11.Why proxy ARP can not support reasonable form of routing?

Q3.                                                                                           (10)

A class C network has to have 10 networks subnetted in which 3 networks contain 50 nodes and others contain less than 14 nodes. Device a subnetting scheme for working in this situation. How supernetting will solve the problem if 20 more such networks are to be added to the very site? Assume 4 class C addresses are available. (Consider complete restructuring) Show in both cases how address resolution by local routers can be made.