Page 147 - DCAP207_NETWORKS_DCAP406_COMPUTER_NETWORKS

Computer Networks/Networks

Notes
Figure 9.3: Routing Method-Distance - Vector Type

There are problems, however, such as:

(1) If exchanging data among routers every 90 seconds, for example, it takes 90 × 10 seconds
that a router detects a problem in a router 10 routers ahead and the route cannot be
changed during this period.
(2) Traffic increases since routing information is continually exchanged.
(3) There is a limit to the maximum amount of routing information (15 for RIP), and routing
is not possible on networks where the number of hops exceeds this maximum.
(4) Metric cost data is only the number of hops, and so selecting the best path is difficult.
However, routing processing is simple, and it is used in small-scale networks in which the
points mentioned above are not a problem. Distance vector routing was used in the ARPANET
routing algorithm and was also used in the Internet under the name RIP. It also found its uses in
early versions of DECnet and Novell’s IPX. AppleTalk and CISCO routers use improved version
of distance vector protocols. In the improved version, each router has a routing table indexed by
and containing one entry for each router in the subnet. This entry has two parts. They are the
preferred outgoing line to use for destination and an estimate of the time or distance to
destination. The metric used is number of hops, time delay in milliseconds and total number of
packets queued along the path or something similar.

Link State Routing

The link state routing is simple. Link state routing algorithm in which each router in the network
learns the network topology then creates a routing table based on this topology. Each router
will send information of its links (Link-State) to its neighbor who will in turn propagate the
information to its neighbors, etc. This occurs until all routers have built a topology of the
network. Each router will then prune the topology, with itself as the root, choosing the least-
cost-path to each router. Thereafter, they build a routing table based on the pruned topology as
shown in Figure 9.4.
The entire topology and delays are measured and distributed to every router. Then Dijkstra’s
algorithm is used to find the shortest path to every other router.

140 LOVELY PROFESSIONAL UNIVERSITY