Network Performance And Network Modelling Computer Science Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

RIP (Routing Information Protocol) is a protocol which is used a dynamic routing and it is used in both the LAN networks and WAN networks. Actually, it is categorized as an interior gateway protocol (IGP). Routing Information Protocol uses the distance-vector routing algorithm. The RIP protocol has been extended for many times, and that is resulting in RIP Version. Both RIP v1 and RIP v2 are still in use up to now, nevertheless, both versions are considered to be make technically out of date by extra advanced techniques like OSPF (Open Shortest Path First) and the OSI protocol IS-IS. RIP has also been modified to be used in IPv6 networks, the standard is known as RIPng (RIP next generation) protocol.

1.0 Background

RIP (Routing Information Protocol) is a managing router information protocol used within an independent network like joint local area network (LAN) or collection of integrated LANs. RIP is categorized by Internet Engineering Task Force (IETF) as an Interior Gateway Protocol (IGP).

Some protocols are using complicated algorithms to determine network distance but RIP will identify the network distance by using a hop count as a way to do that. To find out what is the next host to route a packet for specified destination, every host in the network with a router is going to use the routing table information. RIP is very suitable for small homogeneous networks. In the case of larger networks and more sophisticated networks, RIP's transmission of whole routing table every 30 seconds might be accused a serious mass of more traffic in the network. [1]

2.0 How does RIP work?

By using RIP, the gateway host will send every 30 seconds its whole routing table (that includes all other hosts it already knows about) to its adjacent neighbor host. Continuously, this neighbor host will give the information to its next neighbor and the process will continue until all hosts in the whole network get the same information of routing paths, this situation is called as Network Convergence. [1]

To exchange routing information, RIP uses broadcast User Datagram Protocol (UDP) data packets. Cisco IOS software will send routing information updates every 30 seconds, which is termed as an advertising. If a router does not receive any update from another router for 180 seconds or more, it is going to mark the routes served by the non-updating router as being useless. But if there is still no update after 240 seconds, the router will decide to remove all routing table entries for the non-updating router.

The metric that is used by RIP protocol to evaluate the value of different routes is hop count. The hop count means is the number of routers that can be passing through in a route. A directly connected network will have a metric value of zero; and an unreachable network will get a metric value of 16. So, this small limit of metrics values make RIP an inappropriate routing protocol for large networks.

Any router in the network that is running RIP can receive a default network through an update from another router that is running RIP, or the router itself can supply (generate) the default network itself with RIP. In both situations, the default network will be advertised through RIP to other RIP neighbors.

Cisco IOS software will supply the default network with RIP if one of the following conditions is met:

The ip default-network command is configured.

The default-information originate command is configured.

The default route is learned via another routing protocol or static route and then redistributed into RIP.

RIP will send an updates to all interfaces in the particular networks. If the network of an interface network is not specific, it is not going to be advertised in any RIP update. [2]

3.0 RIP packet types

The RIP protocol identifies two types of packets. These packets could be sent by any device that running the RIP protocol:

3.1 Request packets

A request packet inquires neighboring RIP devices to achieve their distance vector table. The request will specify if the neighbor should return either a specific subset of the table or the whole contents of the table.

3.2 Response packets

A response packet will be sent by a device to advertise the information preserved in its local distance vector table. The table is going to be sent through the following situations:

-The table will be automatically sent every 30 seconds.

-The table will be sent as a response to a request packet produced by another RIP node.

-If triggered updates are supported, then the table will be sent when there is a modification to the local distance vector table.

When a response packet is received by a device, the information included in the update is evaluated against the local distance vector table. If the update includes a lower cost route to a destination, the table will be updated to reproduce the new path. [3]

4.0 RIP packet format

RIP will be used a particular packet format to do sharing information regarding the distances to known network destinations. RIP packets are using UDP datagrams to be transmitted. UDP port 520 is the port used by RIP to send and receive datagrams. The maximum size of the RIP datagrams is about 512 octets. In case if the updates are larger than this size then it must be announced in several datagrams. In the environments, RIP datagrams will be sent by using IP network broadcast address and the MAC all stations broadcast address. But in case of non broadcast and point to point environments, RIP datagrams are in particular addressed to the destination of the device. [3]

5.0 RIP hosts

RIP hosts will be having two types or modes of operation:

5.1 Active mode:

In this mode all devices operating announce their distance vector table and also receive routing updates from other neighboring RIP hosts. Routing devices are usually configured to operate in active mode. [3]

5.2 Passive: (or silent) mode:

In the passive mode all devices operating in just will receive routing updates from other neighboring RIP devices. They do not announce their distance vector table. Actually, the end stations are usually configured to operate in passive mode.

6.0 RIP Routing Metric

To measure the distance between the source and the destination, RIP protocol is used a single routing metric (which is hop count). In a path from the source to the destination, each hop will be assigned a hope count value, which is usually one (1). If the router receives a routing update that includes a new or modified destination network entry, then the router is going to add one (1) to the metric value specified in the update and gets into the network in the routing table. The IP address of sender will be used as the next hop count.

7.0 RIP Stability Features

By implementing a specific on the number of hops permitted in a path travelling from the source to the destination, RIP protocol will prevent routing loops from keep on going forever.The 15 hops is the maximum number in a path. If any router receives a routing update that gets a new or altered entry, and also if increasing the number of metric value by one (1) effects the metric to be unlimited (which is, 16), then the network destination is going to be regarded unreachable. The weakness of this constancy feature is that it restricts the maximum length of RIP network to fewer than 16 hops. RIP protocol consists of number of other strength features that are regular to multiple routing protocols. All these features already designed to supply stability even though possibly quick modifications in the network's topology. For instance, RIP protocol will execute the Split Horizon and Hold-Down methods to avoid wrong routing information from being generated.

8.0 RIP Timers

RIP usually uses various timers to control its performance. These will be involved -flush timer, route-timeout timer, and a routing-update timer route. The function of routing-update timer is to measure the interval among cyclic routing updates. In general, it is set to thirty seconds, with little accidental sum of time will be added each time the timer is reset. Actually all these are done just to help prevent the congestion, which may result when all routers at the same time trying to update their neighbors. Every routing table entry will have a route timeout timer connected with it. If the route timeout expires, then the route will be marked as an expired but is preserved in the table up to the route flush timer expires. [4]

To sum it up, RIP timers are basically four:

Update timer 30s (for broadcast response)

Invalid timer 180s (when does not heard announcements for 180s, does not use)

Hold-down timer 180s (if neighbor increases metric for a network, do not accept directly

Flush 240s timer (after this invalid timer expires, it will mark as unreachable metric 16 until time to flush)

9.0 Differences and Similarities between RIPv1 and RIPv2

The major difference between RIPv1 and RIPv2 is the classless routing. In case of RIPv2, it is going to integrate the addition of the network mask in the update for permitting classless routing advertisements. It is really important for the flexibility required to powerfully make use of network assignments for an ever-decreasing pool of IP addresses.

Of course, there are other dissimilarities, as well. In case of RIPv2, it is used multicast for the destination address for updates, rather than broadcast as in RIPv1. This it will decrease the trouble on the network devices that actually do not need to listen to RIP updates. If as in broadcast, each device on the broadcast domain should as a minimum open the packet of IP and prepares the primary information to decide relevance. But in the multicasting addressing, if any device in the network wants that information, first it is going to listen to that specific address, so if it does not want the RIP information then it does not need to process the multicast address. The RIPv2 is used for the multicast address.

Another thing related to the RIPv2 is authentication. Authentication is used to make sure that routes being spread throughout the network are really coming from approved sources. [5]

The Cisco implementation of RIPv2 supports Message Digest 5 (MD5) and plain text authentications; variable-length subnet masks (VLSMs), CIDR (Classless Inter-Domain Routing) and route summarization. [2]

So we can sum up the differences between RIPv1 & RIPv2 as the following:

9.1 RIPv1

It is a classful distance vector routing protocol.

It does not support discontiguous subnets.

It does not support VLSM.

It does not send subnet mask in routing update.

Its routing updates are broadcast.

It does not use any authentication

9.2 RIPv2

It is a classless distance vector routing protocol that is an enhancement of RIPv1's features.

Its Next hop address is included in updates.

Its routing updates are multicast. ( vs.

Using of authentication is an option [6]

9.3 Similarities between RIPv1 & RIPv2

They use of timers to prevent routing loops.

They use of split horizon or split horizon with poison reverse.

They use of triggered updates.

Maximum number of hop count is 15. [6]

The table 1.0 down shows the summary of differences and similarities between RIPv1 and RIPv2.












Hop count

= 15







Uses Authentication



















Table 1.0: This table shows the differences and similarities between RIPv1 and RIPv2

10.0 RIP limitations

There are a number of limitations observed in RIP environments:

As internets grow, destinations that require a metric of more than 15 become unreachable, making RIP unsuitable for large configurations. On the other hand, if larger metrics were allowed, the convergence of the protocol upon initialization or after topology changes can be lengthy.

The overly simplistic metric leads to suboptimal routing tables, resulting in packets being sent over slow (or otherwise costly) links when better paths are available.

RIP-enabled devices will accept RIP updates from any device. This enables a misconfigure device easily to disrupt (disturb, interrupt) an entire configuration. [7]

Network-demanding table updates: Constant broadcasting of the distance vector table may lead to get result in improved utilization of network resources. As a result this can be affect in reduced-capacity segments.

Fairly slow convergence: RIP, is just like other distance vector protocols, it is quite slow to converge. The algorithms depend on timers to set off routing table advertisements.

There is no support for VLSM (variable length subnet masking) in case of RIPv1; Route advertisements in a RIPv1 situation do not include subnet masking information. This is going to make it not possible for RIP networks to set up VLSM (variable length subnet masks). [3]

11.0 Summary

In spite of RIP's time and the appearance of more complicated routing protocols, RIP is far from outdated. RIP is greatly supported, stable, mature and easy to configure. RIP has a simplicity which is well appropriate for using in the remains networks and also in the small autonomous systems that don't have sufficient plentiful paths to guarantee the overheads of an extra complicated protocol. [3]