Routing protocols and link state explained

Routing protocols play a crucial role in the functioning of computer networks. They determine how data packets travel across networks, ensuring that information reaches its intended destination. In this lesson, I’ll delve into the types of routing protocols available, explore their differences, and discuss their applications. Our primary focus will be on link-state routing protocols, particularly Open Shortest Path First (OSPF).

To start, let us understand the broad categories of routing protocols. In the realm of networking, routing protocols are classified into two main types:

Exterior Gateway Protocol (EGP)
Interior Gateway Protocol (IGP).

The distinction between these two lies in the scope of their operation.

EGP, or Exterior Gateway Protocol, is used for routing between autonomous systems. An autonomous system (AS) refers to a set of routers under the administration of a single entity, such as a company, an Internet Service Provider (ISP), or an organization. EGP facilitates communication between these autonomous systems. The most prominent example of EGP is the Border Gateway Protocol (BGP). BGP is critical to the internet’s infrastructure, acting as the glue that connects autonomous systems and enables global communication. Without BGP, the internet as we know it would cease to exist.

On the other hand, IGP, or Interior Gateway Protocol, operates within an autonomous system. These protocols are used to manage routing within a single network domain, such as a corporate network, a school, or a hospital. Examples of IGPs include Routing Information Protocol (RIP), Open Shortest Path First (OSPF), Intermediate System to Intermediate System (IS-IS), and Enhanced Interior Gateway Routing Protocol (EIGRP). Each of these protocols is suited for different scenarios, depending on the complexity and requirements of the network.

Understanding Interior Gateway Protocols

To better grasp the differences within IGPs, it’s important to examine the two main types of routing mechanisms they employ: distance vector and link-state.

Distance vector protocols, such as RIP, operate on a straightforward principle. Routers share their entire routing table with neighboring routers at regular intervals, typically every 30 seconds. This periodic update mechanism is simple but comes with significant drawbacks. For example, updates are sent as broadcasts (in RIP v1) or multicasts (in RIP v2), which can consume unnecessary bandwidth. Moreover, distance vector protocols provide limited network visibility, as each router only knows about its immediate neighbors and relies on their updates to learn about the rest of the network.

One major challenge of distance vector protocols is the risk of routing loops. These loops can occur when routers circulate incorrect routing information in a larger network, causing inefficiencies and delays. Techniques such as split horizon and route poisoning have been developed to mitigate these issues, but they don’t eliminate the risk entirely.

In contrast, link-state protocols, such as OSPF and IS-IS, offer a more sophisticated approach. Instead of periodic updates, link-state protocols use a triggered update mechanism. When there is a change in the network—such as a link going down or coming up—routers share only the specific change with the rest of the network. This incremental update mechanism reduces bandwidth usage and ensures more efficient communication.

A key advantage of link-state protocols is their ability to provide end-to-end visibility of the network. Routers using OSPF, for example, maintain a complete map of the network topology within their area, allowing them to make more informed and optimal routing decisions. However, this comes at a cost: link-state protocols are more complex to configure and require greater expertise to manage.

Key Differences Between Distance Vector and Link-State Protocols

The distinctions between distance vector and link-state protocols can be summarized in the following table:

Distance Vector	Link State
Entire routing table is sent as an update	Updates are incremental & entire routing table is not sent as an update
Send periodic update (default every 30 sec)	Updates are triggered not periodic
Routers do not have end to end visibility of the entire network	Routers have visibility of the entire network in that area only
Easy to configure	Require expertise knowledge to configure it
Can suffer from Routing Loops	No Routing loops
Example: RIP	Example: OSPF, IS-IS