Open Shortest Path First (OSPF) - 7 : Link State Protocols

This playlist provides an overview of the Open Shortest Path First (OSPF) routing protocol and its advantages and disadvantages:

First we learned that routing protocols are used to exchange information between routers. Then we learned that there are different classes of routing protocols. Now we get to the different algorithms the protocols can use to determine how to forward packets. The two most common types of routing protocols are distance vector protocols and link state protocols - each of these types of protocols uses a different algorithm. RIP uses the distance vector method to forward packets, while OSPF is a link state protocol.

The link state protocol tends to be more complex and CPU intensive than distance vector protocols like RIP. However, link state protocols are superior in the way that they distribute routing information, and in the quality of information distributed.

However, the link state algorithm is simpler than distance vector protocols in one way - link state protocols like OSPF only require that a router establish who its attached neighbors are, and then advertise this information to all other routers. In other words, OSPF routers only have to determine who their direct neighbors are. This is in contrast to RIP updates. RIP updates contain information about how each individual router views the ENTIRE network. So, each router sends a RIP update with information about the whole network, not just its attached devices and subnets.

The concept behind link state algorithms is that a router determines the state of, or status of, its links. Each router constructs a link state advertisement, or LSA, with the status of its links and transmits this to its neighbors. Each router builds a complete list of all routes to all destinations, based on compiling all the LSAs from each router. Convergence is faster because the information goes directly to all routing services.

To sum it up, each router identifies what routers and subnets are directly connected to it. Then, it distributes this information to all other routers. All OSPF routers take all the information and build a table of what the network looks like. Using this table, each router can identify where all subnetworks are located, what routers are in direct connection, and how to get to any specific router.

This provides a complete view for all routers, allowing them to make more intelligent forwarding decisions.