Name:……………………………………….
Instructor’s name:……………………..
Course:……………………………………..
Date:…………………………………………
Network Reconfiguration
Introduction
Routing refers to the process of choosing a path over which to send data from one network to another by forwarding packets over the Internet. There are two types of routing – static and dynamic. In static routing when a packet is to be routed, the static routing table is consulted and the next hop to traverse is decided. Either we have to manually enter the routes or the network administrator can do so. Since it is a manual process, it becomes difficult to work with it when we need to change the routing information frequently or configure a large number of routers.
On the other hand, dynamic routing mechanism heeds to current network configuration, adjusting routing information from time to time. A router first updates all the routes directly connected to the network and then displays the best route to send the packet. Thus, it quickly adapts to equipment failures and network topology changes.
Comparison of routing protocols
Routing protocols can be divided into two groups IGP (Interior Gateway Protocols), and EGP (Exterior Gateway protocols). IGP is used within a single autonomous system having a single network administration and unique routing policy. On the other hand, EGP is used among different autonomous systems having independent administrative entities. (Kar, 2006)
IGP has two categories – distance-vector routing and link-state routing.
Distance-vector routing protocols are:
- Routing Information Protocol (RIP)
- Interior Gateway Routing Protocol (IGRP)
- Enhanced Interior Gateway Routing Protocol (EIGRP)
Link-state routing protocols are:
EGP can be divided into:
- Exterior Gateway Protocol version 3 (EGP3)
- Border Gateway Protocol (BGP)
The RIP is a dynamic routing protocol mainly used in LANs. In terms of bandwidth consumption, every 30 seconds or so, a RIP router broadcasts a lists of networks and subnets it can reach. The maximum number of hops allowed is 15, and the hold down time is 180 seconds.
IGRP reduces bandwidth utilization by sending updates only when some changes occurs in the network. It also converges more quickly than RIP, usually within a second.
EIGRP is an advanced distance-vector routing protocol, that minimizes both the routing instability, which occurs after a topology change and the use of bandwidth, and processing power in the router. Routers supporting EIGRP automatically redistributes route information to IGRP neighbors. It converges quickly and uses minimum bandwidth, which is in kilobits per second.
OSPF routes packets based on the destination IP address found in IP packets. It supports variable-length subnet masking. It can also detect changes in the topology very quickly and converges to a new loop-free routing structure within seconds. Here, individual components of the database are refreshed every 30 minutes, in the absence of topological changes. As the size of the database increases, the amount of link bandwidth used also increases. The size of an OSPF link state database can be very large depending on external LSAs. This requires a huge amount of router memory. Its CPU usage is dominated by the length of time it takes to run the shortest path algorithm. (King, 2006)
IS-IS provides fast convergence, excellent scalability and it is very efficient in its use of network bandwidth.
EGP3 was originally used for interconnecting autonomous systems. It is a simple reach ability protocol, but unlike modern distance-vector and path-vector protocols, it is limited to tree-like topologies. Hence, it is now obsolete.
BGP is one of the cores routing protocols of the Internet. It maintains a table of IP networks among autonomous systems. It makes routing decisions based on paths and network policies. CPU cycles in BGP depend on the stability of the Internet. If the Internet is stable, then the link bandwidth and router CPU cycles used by BGP are due to the exchange of the BGP messages. The frequency of the exchange is 30 seconds.
Distance vector protocols versus link state protocols
In distance vector, routing protocols the routers do not contain information about the whole network topology. It informs other routers about its distance from them and receives similar information from them. It then calculates the network topology and determines the various paths in it. It is mainly used in packet switching and requires that a router periodically inform its neighbors of topology changes.
Link state routing protocols require that router must send the state of its interfaces to every other router in the network. Here each node contains information about the whole network topology and independently calculates the best next hop from it for every other possible destination in the network.
Link state protocols converge quickly, but use more of the routers’ CPU and memory resources. They rely on network events to make topology changes in the network. Compared to link-state protocols, distance-vector routing protocols have less computational complexity and message overhead. However, it has the disadvantage of slow convergence. However, they are simple to handle and are well suited for use with small networks.
Recommended Protocol
For selecting an appropriate routing protocol, we need to consider the following – for a small, simple network, which is not expected to grow we should use a simpler distance vector routing protocol like RIP v2 and, for a large and complex inter-network we should use a newer and more sophisticated link state routing protocol like OSPF. We can also use RIP v2 or OSPF if we want to support variable length subnet masks. Though the outdated RIP v1 is still widely used in private networks, it does not support VLSM and thus is not suited for enterprise networks. (Fletcher, 2007)
Conclusions
Routing protocols certainly have far-reaching effects on the performance and reliability of a network. Thus, we need to choose the best one for our network. All routing protocols have their strengths and weaknesses, and there is no routing protocol, which can be said to be the perfect tool for every network.
References
Fletcher, R; (2007); Network Principals: Beliefs and Knowledge; Believing and Knowing; Dunedin: Howard & Price.
Kar, P; (2006); Networking and Related Applications; Kolkata: Dasgupta & Chatterjee.
King, H; (2006); Software Principals Today; Auckland: HBT & Brooks Ltd.