Apr 1999 NI222: Networks and the Internet (Question 3)

(a)	Explain why we have the redundancy of using both domain name and IP addresses in the Internet.	[3]
	The IP addresses allows the packets to be easily understood by Routers and be forwarded to the next destination. The host and domain name is for the users to easily remember and understand the addresses. Logical alphabet names will definitely be more readable than a 32-bit long number separated by dots. The use of domain name also caters fir the change of IP address of the server without changing the domain name, for example, the server office had move to another state therefore the IP address would definitely change but the name need not, only updating the DNS would be needed.
(b)	What is the DNS? Why does it have to be distributed?	[4]
	DNS stands for Domain Name Service. It is a server within a sub domain that contains all the domain names and IP addresses in a table. All the hosts within its change would request domain name resolution of IP address through it. The DNS has to be a distributed database because the internet is a very large network, it is impractical for a single server to contain all the domain names and IP addresses of the entire network. The volume of traffic for service would also be almost impossible to handle. Therefore the DNS is distributed, a server is allocated for individual domains. If the IP address cannot be found within the Local DNS, the server will request service from the higher level domains or root domains.
(c)	Explain the term port address. Is it the same as an IP address ? Use an example to explain your answer.	[4]
	The port address is the physical address of the connection made where there is communication and sending of data. A server would listen to the service request port, e.g 80 for request for service where then it would use a port to establish a connection. The address range is from 0 onwards. So, the port address is different from IP address. Each application utilizes a port for communication and data transfer, either by default or through the process server which assigns port number to applications. IP address are the internet addresses of the hosts either server or client, it allows the router to navigate the packets to the final destination of the receiving computer. The receiver has several application running simultaneously have the differentiation of application data through the individual ports assigned allows the separation of each application's data.
(d)	Discuss at least two basic differences between IP address and Ethernet address.	[4]
	IP address is usually assigned by the Internet Service Provider, the individual user would not have a fixed IP address, however each user would have a fixed ethernet address that is from their hardware network interface card(NIC). The ethernet address is permanent and is 'burned' onto the NIC, the only way to change ethernet address is to change the NIC. Address Resolution Protocol (ARP) is needed to lookup the host ethernet address. PNS is required for IP address search from domain names. IP Address is of 4 bytes separated by period. Ethernet Address is of 6 bytes separated by hyphen
(e)	Before IP packets can be sent, the destination ethernet address must be determined, using the address resolution protocol (ARP). The first phase of operation of the ARP is to look up the address in a locally-held table. What happens to the IP packets whose address lookup in the ARP table fails? How do the packets eventually get sent?	[5]
	The router would send an ARP to its local network. The ARP consists of 4 fields. Recipient IP and ethernet address, sender IP and ethernet address. It is broadcast out. The node whose IP address is matched will fill in the ethernet address and send the ARP packet back to the router. The router will then update ARP table and proceed to send the packet that it has stored in the Q buffer.