What is a Routing Protocol?
A routing protocol defines how routers communicate with each other, it uses routing algorithms to determine optimal paths and network to transfer data between different network segments on a computer network.
Routing is essential to keep a network running. A successful attack can compromise the router, peering sessions and routing related information. They are used by routers to memorize routes & maintaining the routing tables. A basic understanding of routing protocols will help to identify potential risk and to take protective measures to prevent any cyber attack.
The best practice is to use only one routing protocol throughout the organization. One may manage the network with several routing protocols also coexist to connect two or more ISPs for connectivity. In such a case, the commonly used contract to exchange routes with the service provider is BGP while within the organization OSPF or EIGRP are used.
There are several different possibilities exist while choosing the optimal routing protocol for the organization. Therefore, it is essential to understand the pros and cons of each contract. The various routing protocols can be grouped in several ways. One option is to group them based on whether protocols operate within or between autonomous system. An independent system represents a collection of network devices under one administration. Typical examples of an autonomous system are an internal network of the organization or network infrastructure of an ISP.
Interior Gateway Protocols (IGP) is used within the organization to exchange routes within an autonomous system. They can support small, medium or large sized organizations but with limited scalability. Configuring with very fast convergence is effortless. The most commonly used IGPs are OSPF and EIGRP. RIP is also used, but it is sporadic nowadays. IS-IS is generally found within the ISP internal network.
Exterior Gateway Protocols (EGP) take care of routes between different autonomous systems. BGP is the only EGP that is used as of today. The primary capability of BGP is to exchange a considerable number of routes between different autonomous systems that are part of the Internet.
Distance Vector Protocols |
Link-State Protocols |
Path Vector Protocols |
|
Exchanges routes as vectors of distance and direction:
Distance - Metric of links or hop count
Direction - next-hop neighbor
|
Exchange information about the whole topology:
Each router determines the best path on its own, using SPF
|
Exchange routes as the vector of path and direction:
Path - list of different path attributes
Direction - next-hop neighbor
|
Example - EIGRP and RIPV2 |
Example - OSPF and IS-IS |
Example - BGP |
Distance Vector Protocol determines the distance (such as link metric or number of hops) and direction (vector) to any link in the network. They use routers as a sign along the path to the final destination, the only info that a router knows that about a system is the hop count to reach this network and which interface or way to used to get there. They don’t have a real idea of the network topology. While at first, they support only the periodic exchange of routing information, the most commonly used distance vector protocols are EIGRP and RIPv2, use triggered updates to respond to topology changes.
Difference between RIPv1 and RIPv2 Protocols
RIPv1 Protocol |
RIPv2 Protocol |
It is a Distance Vector routing protocol |
It is basically a Distance Vector routing protocol, which some properties of Link State protocols. |
It is a classful routing protocol, but it doesn't support Variable Length Subnet Masking (VLSM) |
It is a classless routing protocol, which also supports VLSM |
It uses broadcast address 255.255.255.255 to update the routing table in every 30 seconds |
It uses multicast to update the routing table with multicast IP address 224.0.0.9 |
It does not support authentication |
It supports authentication, which helps in confirming that the updates are coming from an authorized source |
Difference between IGRP and EIGRP Protocols
IGRP Protocol |
EIGRP Protocol |
It is a classful routing protocol |
It is a classless routing protocol |
It uses a 24-bit value for bandwidth and delay |
It expresses the metric as a 32-bit value |
It does not support areas or tables, but it promotes multi-part routing |
Routing is done using neighbor tables, topology tables, and routing tables |
The maximum hop count of IGRP routed packets is 255 (default is 100) |
The maximum hop count is 255 |
Link-State Protocol uses the shortest path tree algorithm which creates an abstract of the exact topology of the entire network. Using an analogy of the sign, a link-state routing protocol is like having a complete map of the system. The sign along the way from the source to the destination is not necessary because all link-state routers use an identical map of the network. A router configured with link state protocol uses the link state information to create a topology map and to select the best path to all destination networks in the topology. The examples of link-state routing protocols are OSPF and IS-IS.
Path Vector Protocol exchanges not only information about the existence of destination networks but also the path on how to reach that destination. Path information is used to determine the best routes and to prevent looping. Using an analogy of the sign, routers are not only known with the direction of the destination network, but also with the specific path to the destination. BGP is the most widely used path vector protocol.
Border Gateway Protocol (BGP) is the protocol that enables the communication to happen quickly & efficiently, it is like the postal service on the Internet. For example, when someone drops a letter into a postbox, the post office process that message and chooses an efficient, fast and reliable route to deliver to its recipient. Similarly, BGP is the protocol that makes the Internet work, whenever a user in India loads a website with origin servers in the USA, BGP enables the data routing on the Internet.
Article Link: https://blog.lucideus.com/2019/05/routing-protocols.html