How do all of our different routing protocols and routing methodologies worked together? Well, let's take a look.
We have a dynamic routing tables now. We talked about how, when we have a routing table routing table can either be static or dynamic.
Now a static routing table
is a routing table that we manually enter everything in by hand. So if we have a static routing table or if we have stout, static routes, that means that we went into our routing table on our computer or we went into our routing table on our router. And we told that device.
Okay, if you're trying to reach this location,
enough set. That's all that you do. So that's going to be a static route. That's gonna be your stout, ecstatic routing table.
Our dynamic routing tables function a little bit differently. Are dynamic routing tables automatically find the best route to take or try to find the best route to take?
So how did they do this? Well, how they do this, we break down into a couple different sections,
if our router is trying to find the route for a device or a network inside of our domain or inside of autonomous network. We use I. G P
Far router is trying to find an outside route to, say, a Web server across the world that isn't under our control. Then we're using E g. P.
So right there we break up our main distinction. I g p is internal
and e g p is external,
So the eye for internal e for external
our internal has ah, we can break down our different routing protocols into three different types. We have distance, vector, link, state and hybrid.
Our distance vector routing protocols
are our routing protocols, which are based on how many hops it takes to get from point A to point B.
it takes for us to get from one location to another just essentially, how many stops and other routers we need to make.
So that's what distance vector means. Distance vector means how many hops, how many stops we need to stop and take before we get from one point to another.
We have to. We have three routing protocols that fall under I G P distance vector, and that is ripped
and Whip V two as well as I g R P.
Rip, as we said before, has a max of 15 hops on does not handle class less addresses. Rip V two has more hops, and it does handle class classless addresses. I g R P does not handle classless addresses, but it does have more. It has more hops than it can handle them. Standard rip.
So those were going to be our our dynamic routing table
protocols under I G. P under Distance Vector protocols.
Next, we have our Link State section for R. I. G P protocols and our Link State protocols are going to be protocols which, based off of cost, different costs, such as how round trip time or reliability of a particular device on the network
are different. Empty use. I'll go into the cost of a particular
of a particular link or a particular route
programing protocol, which is going to use links, is going to use a link State is going to use thes metrics, and that's gonna be oh, SPF open shortest path First that's gonna be our most common routing Protocol are most common interior Gateway protocol are most common. I g p
is going to be open shortest path first
and then lastly, we have hybrid. And for hybrid, we hav e i g r p enhanced into your gateway routing protocol.
So our e i g r P is going to be a hybrid of distance, vector and link state.
So it uses both of these metrics
both of these metrics types in order to make its calculations. Um, and it's a Cisco proprietary, just like I g r P is
our hybrid doesn't use
when we say our hybrid protocol. Don't think that our hybrid don't think it eat that. Don't think that e. I. G. R P uses these other protocols in its one protocol. It just uses the different methods of distance, vector and Link state so it uses hop count from distance Vector,
and it also uses costs from links state. And so that's why it's called a hybrid.
Now that's all of our different protocols and different types associated with the I G p
for E g p R X to your Gateway protocol,
we use one. We have one type of protocol, and it's a path vector protocol which means that it's similar to distance vector protocol. It's based on the distance the number of hops it takes to get somewhere. But with e g. P with path vector protocol each of the different domains, each of the different autonomous networks of routers
are going to assign one speaker node
to broadcast out their information.
And this is going to be border gateway protocol. And that's our not only are one path vector protocol, but our one exterior gateway protocol. How do all of these different dynamic riding tables work together? Is it just one router Just determining all these metrics and doing all the calculations by itself?
All of our different routers talk together,
and they say, Hey, this is the information. These were the round tables that I have and they have time stamped. Why their most recent changes? Do you have anything newer? Do you have anything that looks different than my routing table? Let's share information and see if we can come up with a more complete routing table, a better idea of how this whole network is working.
these different vectors are these different. Sorry. These different routing protocols have different tape have different times at which they exchange information in order for the routers to get a better idea of Oh, hey, you have a more up to date table for this entry than I do. Let me let me take that from you and copy it over to my table. And and I have amore
up to date entry here than you do. Let me copy mine over to you.
So that's how are different. Rounding protocols work together and how are different. Routers used these protocols not only to find paths by themselves but also defined paths by sharing information with other routers.
Now, when we do share other information with other routers, we may reach a state called Convergence, also known a steady state.
Convergence is when all of our routers have all of the same topology, information and all of the same network routing information. It's sort of like a we have a peer, our network routing nirvana. It doesn't mean that all of our routes are perfect. It doesn't mean that all of our routes are the best that they could be,
but it means that all of our routers
have the same apology information they all think that our network looks the same. They all think the same thing about what the way our network looks, and they all think the same thing about the way you get to different places. They all have the same routing information.
Convergence is good because it means that are, well, convergence can be good, I'll say, because it means that our routers are talking to each other. It means that they're all caught up with each other, and they have the best idea on how to get where, so that if one person that's connected to a router in the Finance Department
needs to get to a particular server,
and another person who's connected to a router in the sales department needs to get to a particular server. Both of those routers have the same apology and network information not necessarily the same routing tables because
they may have shorter paths to get to this the same server.
But they know they both have the same idea of what their topology looks like. They both have the same. You could say road map as everybody else. They say. The router in the Finance department says Okay, I need to go through routers A, B and C to get there. So
the routers in our finance department
I know that they need to go through
these different routers to get to our server.
And then our router in our sales department knows that it only needs to go to this to server, see,
in order to get to our sales department work to get to our server.
convergence isn't necessarily all the same routing tables because our different routers are on different places on our network, so they'll have slightly different routing tables. They'll have slightly different. Okay, I need to go here to get here, because they may be positioned different. Different locations.
A router that is two hops away from a server needs to go to a different router before it gets to the server than a router that is only one hop away from the server,
so they have different routing tables. They have different places. They need to go in order to get to their destination. But convergence simply means that they all have the same road map
they all share. They all share the same information as far as what their network looks like if they were toe all getting a little off. All your routers were to get in a little router helicopter and go up and take a look at your network apology. They would all have the same idea of what's going on. They would have. They would all have the same network in topology information.
when we're setting up our routers and we're trying to get all routers to communicate, we want to strive for this convergence because it helps with our network flow, and it helps with our traffic flow.
convergence occurs at different rates. This depends on several factors, such as our network size. Larger networks will take larger to achieve convergence than smaller networks. If you have a network with only three routers in it, and you make a change on one of those routers,
thio the topology and you move that router or you move some computers around that are connected to that router.
It'll talk. It'll take a lot less time for that router to talk out to the other two routers and say, Oh, yeah, there's been a change on my routing tables. There's been a change in our networks apology. Let me know. Let me let you know what that is then, if we have an environment with ah 100 routers and there's a change on one of those routers. And that first rounder
tells its three neighbors who tell their three neighbors who tell their three neighbors who tell their three neighbors
until we can eventually get to all 100 routers. Knowing about this change
so larger networks will reach convergence slower than smaller networks. Well,
in theory, there may be a very badly managed, smaller network, which takes large, which takes longer to manage than a larger network. But all of those other factors aside, just based on size larger takes longer to converge on a smaller network
type of routing protocol that we're using the based on the different routing protocols that were using our different routing protocols share information with other routers based on different schedules that they have. So the different routing protocols that were using take different amounts of time to reach convergence
and also hardware and device problems. If we have a heart, if we have a router that's having trouble communicating out, or if we have a jabbering if we have a jabbering router. If we have a router that's going up and coming down or a device that's going coming up and coming down going up and coming down constantly, then we may constantly have
ah swap on our routing table between
Oh, yes, you can. You can go through this route to get there. Oh, wait, no, you can't. Oh, yeah, you can. No, you can't get you can. And these constant changes will take longer to propagate. So if we're having hardware device problems, then I will also take us longer if at all, if at all possible to reach this convergence. So
hopefully hopefully this
chart was a better way of understanding how we use our dynamic routing tables and how each of our different protocols are broken down into either i g p your e g p and how each of those are then additionally broken down into distance vector link state or hybrid or path factor, which is,
take off of distance vector.
And then how all of those different routing protocols are then used for hopefully achieving convergence also known a steady state, our little network routing nirvana in order to make sure that all of our routers have the same apology in the same network. Information