Different dynamic routing protocols build their routing tables in different ways. In this video, you’ll learn about the methods used by link state, distance vector, and hybrid routing protocols.
<< Previous: Static and Dynamic RoutingNext: IGP and EGP >>
When you’re trying to decide what dynamic routing protocol you’d like to use in your environment, you’re going to want to understand how those protocols determine what the next best hop is going to be. So in this video, we’re going to look at the differences between link state routing protocols, distance vector routing protocols, and hybrid routing protocols.
As the name implies, a link state routing protocol is most concerned about the state of the link. Is it up, or is it down? It makes its routing decisions based on connectivity. If it’s up, we can traverse this link and get to that location. If it’s down, we can’t. Now, one of the things that a link state routing protocol might be able to do is to consider how fast the link might be. That way, we can understand not just uptime and availability, but which one is going to be the much faster route to get to that location. And again, that is part of the link itself. That gives us a state of the link. This is pretty scalable, and it’s used a lot in very large networks. The decisions to be able to determine what is the next best hop are done relatively quickly and relatively simply, which makes it easy to calculate and troubleshoot if you’re ever running into a problem with your dynamic routing protocol. A common example of a link state routing protocol might be OSPF or IS-IS. OSPF of course being used inside of many organizations, IS-IS I think is something you will find in very large provider networks. So if you’re wondering, what can I use for my very large network and how can I have a scalable routing protocol, you might want to concentrate on one of those dynamic routing protocols.
A distance vector routing protocol is most concerned about how far away a particular network is. And generally, in routing we think about the distance away based on the number of hops, the number of routers we have to traverse to get to that other location. This is something that requires very little configuration. We don’t have to tell the router how many hops away it is. The dynamic routing protocol is going to communicate between all of the different routers, and the routers will update each other on how far it is to get to a particular location. A distance vector protocol is good if you have a relatively small network. And once you start exceeding a certain number of hops, you’ll find that the protocol doesn’t support the large number of hops or becomes more difficult for the protocol to calculate how many hops away a particular site might be. So if you have a larger network, this may not be the routing protocol for you. Distance vector protocols are commonly ones like RIP, RIPv2, and BGP.
A hybrid routing protocol has some characteristics of a link state routing protocol, and some characteristics of a distance vector routing protocol. An example of a hybrid routing protocol would be something like EIGRP. This is Cisco’s Enhanced Interior Gateway Routing Protocol. And it’s a very powerful routing protocol that can look at many different characteristics. It can look at bandwidth, and load, and delay, and reliability, and MTU, and hop count, and it can use all of those or a number of these to make a decision on what the best route might be to a location. There is some controversy and confusion about EIGRP. Cisco themselves say that this is a distance vector routing protocol. You have other people saying that it’s a hybrid routing protocol. But if you look at all of the different characteristics that are examined, it seems to fit best into this category of a hybrid routing protocol.