Open Shortest Path First (OSPF) - 8 : OSPF

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

So now we get to OSPF itself, which was actually designed to replace RIP. Open Shortest Path First is a routing protocol designed for larger or more complex networks than those normally supported by RIP. The standards for it are covered in RFC 1245 and RFC 1583.
OSPF routing uses a shortest path first, which is another way of saying that it's a link state protocol. As I mentioned, OSPF is also an interior gateway protocol that distributes routing information between routers in a single autonomous system. OSPF chooses the least-cost path as the best path. In RIP routed networks, the least cost between two routers is simply the shortest path or least number of intermediate routers. In OSPF networks, least cost can be identified many ways in addition to RIP's hop count. Using OSPF, least cost can include bandwidth, congestion, monetary costs, etc. The fact that OSPF determines least cost is one of its big advantages compared to RIP, which only considers the distance, or hop count, to the destination.

I mentioned that OSPF is more complex than RIP. Once all routers have constructed their databases based on the LSA information, they run the "shortest path first" algorithm. This results in a tree structure with each router at the "root" of its own tree, and the shortest path to all other destinations mapped out. The selection of the path to these destinations is based on metrics. These metrics can be based on hop count, bandwidth, load, cost, reliability, delay, or they can be controlled statically by the user. This offers the network manager much greater control over how routing occurs in the network.

The advantage of using OSPF in large networks is that it uses areas. By breaking the large network down into smaller domains, traffic is better controlled, and forwarding occurs more efficiently.

Additionally, OSPF is optimized for large complex networks because it provides an equal-cost multi-path routing option, in which packets to a single destination can use separate paths simultaneously.

Finally, the distribution of LSAs is accomplished using a multicast address recognized by all routing services configured for OSPF.

All this sounds really confusing, but here are the basic reasons why OSPF is better than RIP:
1. OSPF networks have quicker convergence time because each router gets network change updates directly from the affected router. RIP requires messages to be passed along from one router to another, to another, to another...
2. OSPF also uses the concept of areas in which local LSAs are kept local and not forwarded to areas not affected
3. OSPF is 'smart' enough to consider multiple options for what constitutes least cost between routers, not just hop count