Routing Information Protocol Is Very Important Computer Science Essay

Published Date: 02 Nov 2017

Aaron Stark

Net 114-102-13SP

George Gibeau

Routing*Information*Protocol

Routing information protocol is very important to internetworking, since it passes information about routes between networks and hosts, allows hosts and gateways to exchange information for the purpose of learning new routes for data to take, and controls the hop count as a routing metric.

RIP was designed to work with moderate sized networks, which used pretty much the same technology. It was not intended to work complex network systems. RIP is widely used for routing traffic in the global Internet and is an interior gateway protocol (IGP), which means that it performs routing within a single autonomous system. On the Internet, an autonomous system is either a single network or a group of networks that are controlled by a common network administrator (or group of administrators) on behalf of a single administrative entity (such as a university, a business enterprise, or a business division). An autonomous system is also sometimes referred to as a routing domain. There are several algorithms to determine routes between networks. One categorization is on the type of information that is needed to be exchanged between the networks. Distance vector algorithms exchange only a small amount of data. Each participating gateway keeps information about all destinations within the network.

RIP is a "distance vector algorithm". The basic distance vector algorithm attempts to find the shortest number of "hops" possible to reach the destination. A hop is whenever you pass through a node. When the network has many interconnected nodes, the shortest distance is harder to find. Distance vector algorithms use tables that contain the "best" routes to every destination in the system. Metrics is used to determine the "best" routes. In small networks, a common metric would be to count how many gateways a message would need to traverse to reach the destination. More complex networks would choose a metric that represents the time delay, cost of sending, or some other quantity. The main point is that the metric represents the "cost" for individual hops. So in the following example you can see that the distance from A to C is only two hops:

A ----- B ----- C

In the diagram just to the right, you can see that the shortest path from A to M could be A-B-C-E-M, A-B-D-H-M, or A-G-J-L-M. There are two algorithms that try to find the shortest path, the Bellman-Ford algorithm and the Dijkstra algorithm. Once one of these algorithms is run, each node in the network will know the shortest path from itself to each other node in the network. Each algorithm uses some sort of fixed metrics for the route comparisons. The fixed metrics can be a time delay, cost, or any other type of value used to compare routes.

Algorithms work based on some assumptions, and that is that the nodes never fail, the cost of "hopping" never changes, you have the space to store all of the data, and that you also have the ability to collect the data. However, in practice, these assumptions are not valid. Nodes can/do crash, cost change, and there is a specified limit on how much information can be kept. This is where protocols come in. Not only do they have to implement the algorithm in code but they need to deal with a changing topology or layout of the network.

The Internet is a collection of networks connected by gateways and RIP is intended for use within the Internet. Gateways are hardware and software that link two different types of networks. They use routes to determine where to send the datagrams. If the destination is on one of the networks directly connected to the host or gateway, it can send the datagram directly to it; otherwise it will try to send the datagram to a gateway that is nearer the destination. Routing is the method of finding a path from a sender to the desired destination, but on the Internet, this would be finding gateways between networks. If the networks are not adjacent, the message must pass through gateways. Once the message gets to the gateway that the destination network is on, the local network router will send the message to its destination.

Each gateway maintains a database, which contains entries of where datagrams should be sent. This database contains a metric attribute that measures the "total distance" to the next gateway. The "total distance" may be the time delay in getting the message to the next gateway, the cost of the sending the message, or a different measurable factor. Each entity or gateway keeps a routing database. There will be an entry for every possible destination in the system. Each destination requires information including an IP address, a gateway, the interface, Metric and a timer. As time goes by, the database will be modified with information about the entities that are directly connected to the system, as well as information from other systems on different networks. However, Split horizons derive from the idea that it is never useful to send information about a route back in the direction from which it came. For the following example keep in mind R1 and R2 represent routers, and N1 and N2 represent networks. The following model demonstrates the split-horizon rule:

N1 ---- R1 ---- N2 ---- R2

Router 1 (R1) initially broadcasts that it has a route to Network 1 (N1). There is no reason for Router 2 (R2) to include this route in its update back to R1 because R1 is closer to N1. The split-horizon rule says that R2 should exclude this route from any updates it sends to R1. The split-horizon rule helps prevent routing loops. Consider, for example, the case where R1's connection to N1 goes down. R2 continues to inform R1 that it can get to N1 (through R1). If R1 does not have sufficient intelligence, it actually might pick up R2's route as an alternative to its failed direct connection, causing a routing loop. Split horizons are implemented because they provide extra stability.

There are a number of limitations of the protocol. It is limited to networks with no more than 15 hops as its longest path; this is to limit the use to moderate sized networks, most RIP networks are flat; meaning in RIP networks, there is no concept of areas or boundaries and Subnet Masks are not supported by RIP version 1. Gateways and lines can fail and come back up and to handle this we are required to update the routes and distribute it throughout the network. The process of establishing routes is called convergence. Transmission of the route through the network takes time, so convergence doesn't happen instantaneously. "Counting to infinity" is a method to help speed-up the convergence of the network. This comes into play when the network system consists of several hundred networks, in which a routing table would involve all of them. The concern is that it would take too much bandwidth and time to build the routing table. This is really only a concern for slower lines. Although modern enhancement have been made to RIP known as RIP version 2. RIP version 2 allows more information to be included in RIP packets and provides a simple authentication mechanism. It expands the amount of useful information carried in RIP datagrams and adds security provisions.

RIP is an efficient routing protocol used in small to moderate sized networks. Its basic algorithm structure is a distance vector algorithm using hop count and other metrics to navigate through the network. RIP has been around since the days of ARPANET and I predict it will be here for quite some time.