Networking And Routing Protocols Essay Example

add and update static routes, which was a challenging task. If a link went down, all routers required manual updates. To handle these complex situations, experts developed routing protocols. Despite various contributors and technology changes in the industry, this overview provides information on the major milestones in router history.

Concept of Routing

Routing entails using a router to guide a packet towards its intended destination.

The router receives a packet from one interface, determines the appropriate interface to forward it to based on routing algorithm and destination address, and then transmits the packet through that interface. To route a packet, the router must satisfy the following conditions:

• The router needs to be aware of the destination address and subnet mask.
• The router must discover neighboring routers in order to establish routes to remote routers.
• The router must identify all possible routes to all remote networks.
• The router needs to determine the optimal path for routing the packet.
• The router must maintain and verify both its routing table and routing information.

In general, there are two types of routing: static and dynamic. Static routing involves manually adding routes to the router table. By default, static routes have an administrative distance of 1.

IP route 172.16.30.0 255.255.255.0 172.16.20.2

Dest n/w subnet mask next hope

Static routing does not impact router CPU or link bandwidth, making it a secure alternative to dynamic routing. However, static routing lacks fault tolerance and requires manual route addition, which can be arduous, especially in a wide area network. Additionally, when the network topology changes or a link fails, the network administrator must manually

update the routes. Despite these drawbacks, static routing can be advantageous in certain scenarios.

Static routing is necessary in stub networks where all traffic is routed towards a gateway. It consumes fewer resources, is easy to configure, more secure, and can handle multiple networks. Default routing is a specific type of static routing that only specifies the exit interface (IP route 0.0.0.0 0.0.0.0 serial1).

The administrative distance for default routing is 0 since it is used to send packets to remote networks when the router lacks information about them in the routing table.

Destination Network Subnet Exit Interface

The next crucial and widely used category in networking is dynamic routing, which will be the main focus of this project.

Dynamic routing involves the real-time updating of the routing table through routing protocols. These protocols exchange information and update the table whenever there are network topology changes, ultimately selecting the best path based on metrics. They play a vital role in updating routing tables in large corporate networks. Additionally, these protocols ensure fault tolerance by broadcasting updates in case of link failures or server shutdowns. Communication rules with neighboring routers are defined by routing protocols for updating router tables.

Dynamic protocols are more reliable and robust in networks, particularly large-scale ones, although they consume more CPU power and bandwidth. The rules determine the algorithm and method for exchanging information between neighbors. Routing protocols can be categorized based on their characteristics, with one way being to divide them into routing and routed protocols.

Routed protocols, such as TCP/IP, IPX/SPX, and AppleTalk, are responsible for handling the actual transfer of data. On the other hand, routing protocols like RIP, RIP v.2, IGRP, EIGRP, OSPF, BGP

and others are used to exchange routing information between routers. Dynamic protocols can be further categorized into:

• Interior Gateway Protocols (IGP) and Exterior Gateway Protocols (EGP)
• Classful and Classless
• Distance Vector, Link-state and Hybrid protocols

The classification of IGP and EGP is based on autonomous systems. An autonomous system (AS) refers to a collection of networks within a single administrative domain.

IGP protocols (such as RIP, RIP v.2, IGRP, EGRP, OSPF, and IS-IS) are used to exchange router information within the same AS number, while EGP (such as BGP) is used to exchange router information between different AS numbers. Class-full routing protocols do not advertise the subnet mask but only the class-full address in their advertisements. On the other hand, class-less protocols advertise the subnet mask as well. RIP and IGRP are examples of class-full protocols, while RIP v.2, EIGRP, OSPF, and IS-IS are examples of class-less protocols. Another important distinction among these protocols is that they can be categorized as Distance Vector, Link State, or Hybrid.

• Distance vector protocols
• Advertise periodically
• Advertise full routing table
• Advertise only for directly connected routers
• High convergence time
• Limited number of hops
• Suffer from routing loop
• Do not establish neighbor relationship
• Protocols – RIP, IGRP
• Link state protocols
• Advertise only when network triggered
• Advertise only the update
• Flood the advertisement
• Convergence is low
• No limits in hop count and suitable for large network
• No routing loops
• Establish neighbor relation in formal way
• Protocols
  

– OSPF & IS-IS

It’s a combination of both Distance vector and Link-state. EIGRP share such routing characteristics.

Dynamic routing Protocols

Routing Information Protocol (RIPv1)

Routing information protocol version 1 known as RIP is the initial routing protocol to be implemented in ARPANET in 1967. As classified before RIP is a class-full, distance vector and interior gateway protocol (IGP).

RIP was created using the Bellman-Ford algorithm and utilizes hop count as the metric value. It determines the best path based on the lowest hop count. In order to prevent routing loops and maintain stability, RIP restricts the number of hosts supported in a network. A maximum of 15 hops are permitted in a network, with the 16th hop designated as having an infinite administrative distance, rendering it unreachable and un-shareable. Updates between routers are transmitted using the broadcast address 255.255.255.255.

This issue arises when all the routers attempt to send updates every 30 seconds simultaneously, causing a strain on the bandwidth due to their synchronization. To address this, a random time delay is added when resetting the timer, in addition to the 30 seconds, in order to prevent congestion. The route timeout timer determines how long a record remains valid before receiving an update with the same record. If the router does not receive an update within this timeframe, it

marks the record for deletion and holds it until the flush time expires. Once the flush time has expired, the record will be permanently removed from the table. The Rip protocol maintains stability by limiting the number of hops to prevent the propagation of routing loops.

RIP utilizes split horizon, route poisoning, and timing mechanisms to prevent the spread of incorrect information. However, the restriction on the number of hops becomes a drawback in large-scale networks. Another limitation of RIP is that it only supports classful advertising. Additionally, there is a security concern with version 1 as routing updates cannot be authenticated. Despite being an older protocol, RIP is still in use in routers due to its ease of configuration, stability, suitability for stub networks, and widespread adoption.

The Routing Information Protocol (RIPV2)

The Routing Information Protocol (RIP) version 2 was standardized and released in 1993 to address the lack of important features in version 1. Version 2 introduced enhancements for variable length subnet masking (VLSM) and classless routing with subnet masks, which were critical updates from the previous version. Version 2 updates carry more information and have simple authentication enabled. Updates sent using RIPV2 utilize the multicast address 224.0.0.9 to ensure that only hosts involved in routing receive the updates.

This version also maintains the maximum number of hops to 15.

Open Shortest Path First (OSPF) Routing Protocol

Open shortest past first (OSPF) plays a key role in IP networks for several reasons. It was drafted to be used with the internet protocol suite with high functionality as a non proprietary protocol. OSPF is an interior gateway routing

protocol which routes packets between the same autonomous systems. It has an administrative distance of 110.

The protocol is designed to fully support VLSM (Variable Length Subnet Masking) or CIDR (Classless Inter-Domain Routing). It also supports manual summarized advertisement. It adopts a link state protocol approach, ensuring scalability, quick convergence, and loop-free routing. In the event of a topology change or link failure, it rapidly converges to establish a new loop-free route.

The metric value is determined by calculating the cost. The best path is found using the Dijkstra algorithm for calculating the shortest path. OSPF sends updates using multicast addresses, specifically 224.0.0.5 for sending updates and 224.0.0.6 for receiving updates.

OSPF maintains three types of tables: the routing table, neighbor table, and database table. It uses the Hello protocol to establish a neighbor relationship and maintain a neighbor table. The attributes of the Hello protocol are the router ID, priority (default 1), hello interval (10 sec), dead interval (40 sec), authentication bit, and stub area flag.

The relationship is established based on the router ID. To establish a neighbor relationship, the timers (hello & dead), network mask, area ID, and authentication password should be the same.

The use of area is employed in OSPF to facilitate communication between routers. OSPF can be set up as either a single area or a multi-area network. The purpose of areas is to control the dissemination of updates within a single area. In order to limit the updates and calculations for finding the best path using Dijkstra's algorithm to one area, an OSPF domain is divided into areas and assigned 32-bit identifiers. It is important to meticulously plan and configure

areas in order to organize hosts and routers into logical groupings.

Each area has its own link state database, which is distributed to other networks through a connecting router. This design minimizes traffic between areas and keeps the topology confidential to other areas. In OSPF with a single area, all interfaces within that network belong to the same network. The diagram below illustrates a configuration in OSPF with a single area.

Multi-area OSPF requires all other areas to directly or virtually connect to the backbone area (area 0). The provided diagram serves as an example of a multi-area configuration. A multiple area OSPF must have at least one backbone/zero area and can have multiple non-backbones. The zero area serves as the core area for all other areas, with all other areas connecting to it to receive updates. Additionally, OSPF permits the configuration of stub networks.

In OSPF stub networks, external updates are not flooded into the stub area, reducing the database and memory consumption. When a stub network area is configured, default routing is used to connect to external areas. OSPF defines the following router states:

• Area border router (ABR)
• Autonomous system boundary router (ASBR)
• Internal router (IR)
• Backbone router (BR)

Routers in an OSPF network can have one or more of these roles. The router identifier must be defined in a dotted decimal format to associate each OSPF instance with an ID.

If no explicit specification is made, the router ID will be assigned to the highest logical IP. The area border router (ABR) is a common router that is placed at the edge of the backbone area in order to connect other areas via

its interfaces. The ABR stores copies of both the backbone and connected area link state databases in its memory. The autonomous system boundary router (ASBR) is responsible for connecting an autonomous system to a non-OSPF network. The ASBR acts as a gateway for connecting an AS to other routing protocol networks such as EIGRP, RIP, BGP, static, etc. It is also used to exchange routes that it has learned from other AS numbers through its own AS number.

The router that has all its interfaces and neighbor relationship within an area is known as an Internal Router (IR). All routers that are part of the backbone area are referred to as backbone routers (BR). ABR, which stands for Area Border Router, is also considered a BR as it is connected to the backbone via a physical or logical link. In OSPF configurations, routers elect a designated router (DR) and a backup designated router (BDR). The designated router (DR) is elected on a multi-access network segment to facilitate the exchange of routing information with other routers.

The role of the designated router (DR) is to multicast the router update it receives to the other routers, eliminating the need for broadcasting. Instead of listening to multiple routers for routing updates, the other routers only listen to the DR. This one-to-many approach ensures that updates are only sent to the DR, which in turn updates all the routers within the segment. This election mechanism greatly reduces network traffic.

The router with the highest priority among the routers will be chosen as the Designated Router. If multiple routers have the same priority, the Router ID will be used to break the tie.

In multi-access networks, the Backup Designated Router (BDR) will be elected next. The BDR serves as a standby router for the DR in case the DR is no longer available. The second router in the election process will become the BDR. If both routers become unavailable, the election process will be repeated.

The Border Gateway Router (BDR) does not multicast updates received from adjacent routers. Instead, OSPF adjacency is used to directly share routing updates between routers. The establishment of adjacency relies on the OSPF configuration of the routers. From an OSPF configuration perspective, networks can be classified as Broadcast multi-access networks, where routers have direct access to all other routers through direct links.

Broadcast multi-access refers to a network where multiple devices, such as Ethernet and Token ring, can access the same network. In this type of network, when an OSPF packet is sent, it is broadcasted to all routers on the network. This is why in an OSPF broadcast multi-access network, a designated router (DR) and a backup designated router (BDR) need to be elected.

The Non-broadcast Multi Access (NBMA) network enables data transmission between hosts in the network either over a virtual link or across a switching device. Examples of NBMA networks include X.25, ATM, and Frame relay. In NBMA, all devices are connected through a shared medium, but it does not support broadcast or multicast capabilities.

Instead, OSPF sends the hello packet to each router in the network individually, requiring special configuration and proper specification of neighbor relationships. Power Line Communication (PLC) is also classified as a Non-broadcast Multiple Access network. Point-to-point connections involve both routers' endpoints being directly connected, enabling a single path for

communication. High-Level Data Link Control (HDLC) and Point-to-Point Protocol (PPP) serve as potential examples of P2P.

In point-to-point networks, the connection between endpoints can either be a direct serial cable or a virtual link spanning a larger distance between two routers. However, in both cases, the need for electing a designated router (DR) and backup designated router (BDR) is eliminated when implementing OSPF. The neighbors are automatically identified in a point-to-point setup. Point-to-multipoint topology, on the other hand, involves connecting a single interface of a router to multiple destination routers. All devices in point-to-multipoint are part of the same network.

Conventionally, routers in a broadcast network could automatically identify their neighbors.

Enhanced Interior Gateway Routing Protocol (EIGRP)

Enhanced Interior Gateway Routing Protocol (EIGRP) is a proprietary, hybrid protocol owned by Cisco. It was developed by CISCO as a successor to IGRP. Although it is not a version of IGRP, it is completely distinct. It functions as both a link state and distance vector protocol. It is also a classless protocol.

The administrative distance for EIGRP is 90 and it utilizes a distinct algorithm called the Diffusing Update Algorithm (DUAL). The DUAL algorithm guarantees swift convergence and loop-free routing for optimal path selection. Additionally, EIGRP also supports unequal cost balancing.

The use of multicast address 224.0.0.0 and autonomous system number is employed by EIGRP for sending updates. It keeps three types of tables: Neighbor table, Topology table, and Hello packets with timers to maintain precise data about directly connected and accessible neighboring routers. The Topology table aggregates all advertised routes with adjoining metrics for all destinations advertised by neighbor routers.

Furthermore, during the aggregation process, a successor and a feasible successor will

be determined and saved. The successor path represents the optimal route to reach a destination, considering the lowest total advertised distance from a neighboring router and the distance required to reach that neighbor. This route will be configured on the router. Additionally, a feasible successor, which has a higher metric than the successor, may be considered as the next potential successor.

The route is not installed but is kept in the topology table as an alternative. When the primary route becomes unavailable, the router will automatically add the feasible route as a successor. In the table, the state of a route for a destination can be marked as either active or passive. If the router finds that the successor is unavailable and there are no backup routes, it will query the neighbor routers. This state is known as active, and it will change to passive once it receives a reply.

The purpose of this process is to create a path without loops for destinations. The routing table contains the actual routes for all destinations, which are derived from the previous calculation of the topology table. This table includes a successor route and, if available, an optional feasible route.

Network Modelling

Network modelling is a key concept in the deployment, planning, design, and implementation of networks.

Modelling is a necessary step in project development that helps describe the project's concept. Before creating a network model, it is important to define network analysis and network designing. The network analysis phase involves identifying requirements, objectives, and problem areas. It also includes describing the router and routing concepts, including how routing protocols are used to route packets and how to configure them.

The implementation stage

considers all requirements, fulfilling them accordingly. In the design part, we define the appropriate network deployment, aided by the use of network modelling. Network modelling provides valuable assistance in generating ideas and creating the best possible network model. Therefore, for this project, I have chosen the OPNET simulator to create network models.

OPNET Modeller 15.0 (Optimized Network Engineering Tools)

Currently, OPNET is regarded as one of the top network modelling tools in the field of network technologies. It offers the capability to design network models using a wide range of network equipments. By utilizing OPNET, networking designers are able to gain a deeper understanding of the design phase before undertaking the development process. This ultimately leads to a reduction in both time and cost associated with prototyping hardware equipments.

We are able to analyze and measure the performance and behavior of the proposed Model system through event simulations. The OPNET tool has several features, including three main editors: Project Editor, Node Editor, and Link Editor. The Project Editor provides a graphical interface for network topology nodes such as subnets, hubs, switches, routers, etc., and various types of links to facilitate communication among these devices. All of these elements are designed with a user-friendly graphical interface to make it easy for end users. The Node Editor provides a clear picture of the internal architecture of the nodes by investigating the data flow between relevant nodes.

Node model is capable of sending, receiving, and creating network traffic with other node model through the use of packets.

Process Editor

The Process Editor discusses the creation of processes and events that result from implementing specific process operations on the network.

This includes understanding the behavior and functionality of the node model. During simulation, each node model can generate processes for various events, providing information on their state and functionality. It is important to note that simulated networks cannot be directly compared to real-world traffic in terms of timing. However, they do offer insights into factors such as required bandwidth, potential jamming points, and strategies for problem avoidance.