So we've talked about some of our different routing protocol major classifications such as our distance vector, our link state in our hybrid. But what is our I g p versus IGP? Well,
we have another classification that we break down our routing protocols into. And this is an interior gateway protocol or an exterior Gateway Protocol i g p Interior Gateway and IGP exterior Gateway
IVP Protocols. I g P Routing protocols are protocols which worked inside an autonomous network, which means they work inside a network which is singling managed, and Singley connected with all of our different routers. So
the Internet is not an autonomous network
we aren't using rip or with V two r o S P F. In order to connect all the routers across the world across the Internet. That's not the way that works.
But if we have, say, a large enterprise building or a large government building or a office building that we're using and trying to get multiple different routers to communicate with each other than those routers are communicating and those routers air setting up their routing tables using an I g. P using an interior gateway protocol,
we think interior. We think of small and we think of internal networks
rather than external giant public networks.
Now e g p Exterior Gateway Protocol is
connecting out to the Internet is connecting out multiple large giant networks together
and R E G P protocols are exterior gateway protocols. There's only one that we use and this is border gateway protocol be GP, whereas I g p uses Rip rip v 20 S P f i g r p or E G R P.
Exterior Gateway protocol only uses be GP only uses border gateway protocol.
So what is border Gateway Protocol? Border gateway protocol is a path vector protocol and path vector.
You guys haven't heard it yet because I haven't mentioned it. Well, you haven't heard of it yet. In this video segment path Vector protocol is similar to our distance vector, which we mentioned. Our distance vector is
where we measured based on hops between different routers. Path factor is similar to distance vector, except we only have one speaker node per domain of routers.
So think of it this way. If if our border gateway protocol
I needed to talk to every single router in the world in order to determine the paths.
It would be a very, very difficult. It would be very, very difficult for that border gateway protocol to ever come to any sort of conclusion as to what was the best path. But with our border gateway protocol,
if we have our router here,
and this router is going to try to talk to other routers in order to get information on what they're routing tables look like And what information is different metrics that they're trying to use? If we have this, if we have one router
So we have our one main router here that connects to all the other routers back in our domain
and then over out on the Internet were connecting to domain, which has one route one main router, which connects the multiple different routers
rather than all of them all. All of the routers across our border gate will weigh protocol talking to each other
in each of our domains. We choose one speaker note. We choose one representative, think of each state chooses their senators. Each domain chooses its speaker node.
So we choose our one main speaker node
to talk to the other speaker nodes. So this greatly reduces on the traffic that we're sending out over the exterior. Um, this could help us to manage what kind of information we're sending out over the exterior because they're different. Route of routing protocols not only transmit different metric information,
but they also transmit rounding tables. They let other routers no.
Okay. Hey, these are the routing tables I've got. Let me see the writing tables that you've got. And if we have our network connected to the Internet,
we don't necessarily want everyone on the Internet to be able to see our routing tables. Why not? What's the harm? Well, there could be potentially a lot of harm.
one layer in all of our different security layers is layer by obscurity. Now, this is not
the best layer of security is not the strongest layer of security, but it is a still very viable additional layer of security, the tack on to all of our layers.
What security by obscurity means other than just rhyming means that we are helping secure our networks by not broadcasting out to everybody. What's inside our network.
It's harder for Attackers, and it takes more time for Attackers to be able to do what's called a numerator. Our network and they enumerates our network. They find what's in our network if we don't broadcast out. Hey, here's my rounding tables. That list
what, dummy? What servers I have on my network? What? This name server, How how you get there
and what routers you need to go through in order to get to my work stations and to get to my Web server and to get to this and that and the other. Because remember, our routing tables were gonna include information such as are different I P addresses and sub net masks and a lot of information that we just may not want
computers outside of our network outside of our organization to know.
So we can limit that by limiting what information is being broadcasted out from the speaker note for our domain on our border gateway protocol. So
again, I g p interior gateway protocol is going to be inside autonomous network, which is going to be all of our protocols we have over here on our left
and E g. P r. Exterior Gateway protocol is going to be connecting to our external networks and routers, and it's going to utilize
be GP border Gateway Protocol on Border Gateway Protocol is the only X two year gateway protocol that we use and border Gateway Protocol is a path vector protocol, which means it's similar to a distance vector, except we only have one speaker node for domains.
Now we have all of our GP protocols, all of our interior gateway protocols. These are different protocols that are routers are going to be using in our environment. All of our different routers on all of our different floors and offices across our building across our domain are going to be using in order to communicate with each other in order to
exchange routing table information in exchange, different vector path information and exchange speeds, and Leighton see and all that
they're going to use an I. G. P and into your gateway protocol.
We have five into your gateway protocols listed up here.
Rip rip version, too.
i g r p and e I g r p. Don't confuse i g r p with a GP or E i g r p
e g p. Your eye GP um, I G R P
just one of the five protocols it's not. In addition, it's not an additional
that's not like an additional type of protocol or anything like that, just because it has i g r P so know the difference between a GP and I G R P.
So rip say, is for routing information protocol.
Rip is a distance vector protocol, which means it's based on hop count. But rip has a lot of limitations.
Our routers that air using rip in order to communicate in order to determine which routes they should take have can handle no more than 15 hops,
which in a small network with a couple of routers isn't a big deal. But in a larger network with a lot more routers is a big deal.
Um, so our routers that our routers that air communicating and that are finding paths using rip are going to be using just hop count in their analysis will set static routes. But they're dynamic routes are just going to be hot counts because remember,
all of our protocols were talking about right now are routed. Our dynamic routing table protocols are protocols which are used in orderto update our dynamic tables on our route on our routers.
So Rip is on. Lee is only going to utilize hop count,
but it's not going to be ableto handle routes with more than 15 hops.
It's also not going to allow us to use classless I p addressing It's not gonna be ableto handle. What we refer to is submitting.
if we're trying to set up classless I p addressing if we're trying to use set up different sub nets on our network, rip routers, which are using rip, aren't going to be able to handle that. They can't handle classless. Addressing that can only handle class full addressing, which is our standard Class A
2 55.0 which is our standard sub nets class 8 to 55 000
Class B 2 55 to 55 00 and classy to 55 to 55 to 55 0
essentially are ripped.
Protocol just looks at the I P address and says, Oh, this I p address has a standard has a default sub net mask of this. So that's what it's gonna be.
It doesn't it doesn't take a look at this. It doesn't need to take a look at the subnet mask. It doesn't take a look at the cider notation. It just says, Oh, this I p address needs this sub net mask. I'm gonna apply that, and I'm gonna look for networks. I'm gonna look for routers, which managed that. So we're trying to set up classless addressing It's not gonna work in this environment with We're just using rip.
rip V two routing information protocol version to which helped, which goes into address some of these issues. Now, which version two is still a distance vector protocol, Which means it's still just using hop counts. So still a little bit weak.
V two, however, has support form or hops. So gives us significantly more significantly more hops than 15 as our limit.
And it will also give us support for classless addressing. So we'll take a look at that 70 mask and say, Wait, this isn't a default sub net mask. So this must mean that we're trying to do some sub netting here, so it ripped V to if if you're using one of the rips than its
you wanted to be the rip. If you have to use one of the routing information protocols, you wanted to be the
ripped V two because Rip V two is gonna give you support from more hops and support for classless addresses.
Then we have O S P F O S P F stands for open shortest path first.
Now oh, SPF is the most commonly used interior gateway protocol. It's the most common protocol that you'll find on routers using them, using it to find different routes across networks. So remember, oh, SPF open shortest palate path First,
it's the most commonly used interior gateway protocol I g p for finding routes and setting up routing tables on dynamic routing tables.
Oh, SPF uses link state analysis. Remember, we mentioned that links state analysis is
checks out things such as how long it's taking to get from one location to another, the round trip time it checks out things like the band with our empty you using that cost based analysis of Okay, if I'm gonna send this packet from my location to this location using this route, what's the cost of that?
Is that gonna be a lower cost.
Ah, lower round trip time. A better A better connection back to me, which is a lower cost, which is good. Or is it going to be a higher cost? Which is bad? Bad, Too much, Which is bad. Bad route for me to use? That's a band multiple times there. Um so high cost is bad. Low cost is good.
So that's what links stayed in out. That's what links data analysis is we talked about that earlier, though like ST analysis is what is going to be used by open shortest path. First Wes pf are most commonly used into your gateway protocol.
And then lastly, we have will last two. We have I g r P and E I G r P
I g r P and E I g R p r both
on Cisco routers. They are
Cisco protocols, so you'll see them there. I g r P stands for interior gateway routing protocol, so
I don't get it confused with standard I g p I G p Interior Gateway Protocol contains all of these other protocols,
so I g. R P is a Cisco distance vector protocol which means again. Distance vector, just like Rip and Rip V two is hot based.
However, I G r P has more hot can handle more hops than rip,
no classless, so it can't handle classless addresses. Forgot to mention O S p f can handle. It does handle classless addresses so it can handle some netting.
I g r P cannot and rip cannot handle classless addresses, but I g r P can handle more hops than rip camp.
just like whip needed something to fix this problem with not handling classless addresses. Uh, we released E I G R P, which is enhanced interior gateway routing protocol enhanced I g R P.
This is a Cisco hybrid
protocol, which means that we have our enhancements, the idea, or pee in that it not only uses hop counts, but it also uses link state analysis. So it also uses cost based analysis. And he uses, uh, uses a combination of hop count
and cost to each each link in order to determine which route are routes or the best.
So our e i G. R P is also going to be able to support classless. Hey, so
that's gonna be our five routing protocols. Rip routing Information Protocol Distance vector No more than 15 hops, No classless
rip Be two more hops than rip one and has support for classless still addict. This inspector
oh, SPF opened shortest past first is going to be the most used i g p
link State analysis So is cost based
i g r p and E i g R P r Both Cisco Interior Gateway Routing protocol Enhanced interior Gateway Routing protocol
I g r p is a distance vector does not handle classless
e i g r p is a hybrid and it does handle classless.
using our knowledge now of our i g p versus r e g p
were ableto recognize how we make our distinction between our interior gateway routing protocols,
our interior Gately routing protocols being again
inside of our network
and then e i g in the e g p exterior gateway protocol being
our network's connecting out to all of the different
different domains. All those different networks out there.
GP utilizes B g p border Gateway protocol in which each domain chooses a speaker. Note
one speaker note. Part domain
and I g p contains all of these different several protocols, which we talked about.
So hopefully you'll be able to, um, take these and we'll be able to pick the best routing protocol for our routers, and we'll be able to find out how able to understand more about how our routers actually find their paths using these different protocols.