August
1999 QUESTION 5 Total Marks: 20 Marks |
Click here to access other
questions
SUGGESTED SOLUTIONS |
| (a) | Distinguish between static and dynamic routing strategies. Explain the advantages of each strategy. | [4] |
| Static routing: All
routes between all possible nodes in a network are computed before the network is up and
running. It is very efficient, there is no overhead of making decisions at routers. Dynamic routing: Routes are recomputed and configured during the operation of the network. It can redirect the packets by reacting to changing network traffic patterns.
|
||
| (b) | Describe the flood routing technique including its advantages and disadvantages. Also explain some circumstances in which flood routing might be useful. | [6] |
| Router forward packets
to all ports except the incoming port by generating large number of duplicate packets.
Advantage is every destination in the network is reachable. Disadvantage is a waster of
link capacity, loops must be detected or prevented.. Loops are prevented by using a hop
count decremented each time a packet is forwarded; no forwarding of a packet with hop
count of zero. Loops can be detected by using sequence numbers (per host), and having
routers remember which packets they have seen. Flood routing is not practical in general. Useful in military applications, where large number of nodes (or links) get destroyed any minute. Also useful in any application where robustness is important.
|
||
| (c) | In the internet's hierarchical routing, exterior and interior protocols are used. Distinguish between these two, using suitable examples. | [4] |
| Exterior Gateway
Protocol: Used for routing information between the autonomous systems (AS) at the highest
level of Internet. Example, Border Gateway Protocol version 4 (BGPv4). Interior Gateway Protocol: Used for routing information within the autonomous systems (AS). Example, Open Shortest Path First (OSPF).
|
||
| (d) | Version 4 of the internet protocol (IP) uses 32-bit addresses. In theory 32 bits could generate over 4 billion IP address, but the world is running out of addresses. Explain why this is the case, using examples. | [2] |
| The IPv4 address
formats means that only a fraction of the address space can be used. For example, the
class D addresses, which are used for multicasting, occupy 0.5 billion addresses.
|
||
| (e) | IP version 6 (IPv6) uses 128-bit addresses. When telephone companies change the numbering scheme, everyone is given an instruction like 'add 2 to the beginning of each phone number'. Will the majority of internet users notice when their service provider starts using IPv6? Why/why not? | [4] |
| IPv6 has been designed
to allow an easy phasing-in, and includes support to automatically renumber entire
networks. Layered protocols mean that most users are not aware of IP addresses, because
they use DNS services, so most users will not notice. A software upgrade will be necessary at some point.
|