Routing is one of the most important concepts of networking. It is actually a process of paths selection in a network via which the traffic can be sent. Routing can be carried out for a number of different kinds of networks. These networks might include the following:
– Telephone network i.e., the circuit switching
– Electronic data networks (for example, internet)
– The transportation networks
The primary concern of this article is routing in electronic data networks.
Packet Switching Technology
We consider here the packet switching technology for routing.
– In this technology, the packet forwarding is directed through the routing process.
– The logically addressed packets are transitioned from the source towards destination through the following which are actually the immediate nodes:
d) Switches and
e) The firewalls
– Routing and packet forwarding processes can also be performed by the general purpose computers even though they do not have any specialized hardware for this.
– This may cause a limited performance in them.
– In routing, the packet forwarding process is directed with the help of tables called the ‘routing tables’.
– These tables keep a record regarding the various routes that can be taken to the network destinations.
– Therefore, the construction of the routing tables is very important.
– The tables are stored in the memory of the routing devices called the ‘routers’.
– Use of these tables makes an efficient routing process.
– There are algorithms that are involved in the routing process called the routing algorithms.
– A single network path might be used by most of the algorithms at the same time.
– But multiple alternative paths can be utilized with the help of the so called ‘multi – path routing techniques’.
– When it comes to the case of equal and overlapping routes, it gets difficult to decide which route should be entered in to the routing table.
– The following elements are considered:
a) Prefix – length: This gives preference to the longer sub net masks. This element is independent of the fact whether it is used within same protocol or in different protocols.
b) Metric: This gives preference to a lower cost or metric. This element can be used only within one and the same routing protocol.
c) Administrative distance: This gives preference to a lower distance. This element can be used only within the case of different routing protocols.
Routing and Bridging
– Routing is quite a many times contrasted with what is called the bridging.
– This is so because the network structures are assumed to be structured and that proximity is because of the similar addresses within a network.
– Only a single routing table entry is allowed in the routing table in structured addresses for representing the route to other devices.
– In a narrow sense, the structured addressing can be also called as the routing.
– It has come to outperform the bridging or the unstructured addressing that was mostly used with the large networks. – It is now known the most dominant form of addressing.
– However, bridging is still prominent in the local networks.
– Today there are a number of routing schemes available:
a) Unicast: Message is sent to one specified node.
b) Broadcast: Message is sent to all the nodes.
c) Multicast: Message is sent to few nodes that to whom the message is of some interest.
d) Anycast: Message is sent to anyone of a group of nodes that is nearest to the source.
e) Geocast: Message is sent to the whole geographic area.
– Out of all these schemes, the unicast is most prominent.
– In non – adaptive routing or the static routing, the configuration of the routing tables is done manually.
– But this is not feasible in large networks that have very complex topologies.
– There topologies keep changing quite frequently because of which manual configuration of the routing tables for them is not possible.
– Adaptive routing or the dynamic provides a solution for this.
– Here, the tables are created automatically.