Join Mark Jacob for an in-depth discussion in this video Introduction to network topologies, part of Learning Networking.
Let's say you're finally getting around to the point where you're gonna put your network together. Disregarding the fact that it probably already is put together. Nevertheless, if you're gonna put your network together, one of the things you need to understand is network topology. What network topology's gonna be in play. And we're gonna talk about three main ones today. Bus, ring, star topology. An important distinction to be made when we're talking about the topic of topology is that there is a physical topology and a logical topology.
To me physical topology is kind of if I'm flying over my company's network in a balloon. And somebody rips the top off of the building. I see it, I see where the cables are. I see where the equipment is. I see the physical layout of the network. But that has very little to do with the flow of network data. That's the logical part of it. In other words, how does the data, how do the data flow through the network? So let's talk about the various topology solutions.
The first of which is called a bus topology. If you think back to the early days of networking. If you go back far enough, you can think back to the days of actually the thicknet and the vampire tap. Where if you were connecting a node or a workstation, you actually screwed in the vampire tap. Ding, ding, ding, you're connected. And it was this cable. And of course, all the nodes were connected to the cable. Well, this is very much like this idea of a bus topology. And I kind of have a mindset or a mental picture of if you're driving an actual bus.
You're driving down a street, and there's bus stops. People get on, people get off. That's the bus. So the flow of data on this bus. This, of course, here representing the bus. And if you think about, for instance, if this workstation here, this node will send, transmit some data. Guess what? Everybody else is gonna hear it. Regardless of whether, let's say there's a file transfer going on between these two machines. Because it's a bus topology, everybody else on this same bus is having to process that information.
At least far enough up their OSI or TCP/IP stack to decide if this if for me. And if it's not, gonna throw it away. But again, they're spending time looking at it. This, like I say, is just a quick introduction to the idea of a bus topology. Single cable, everybody plugged into it. You want to advance to ring. And let's have our, let's do this. Let's make some little nodes here that are connected.
And a ring topology would be very common or more common in a fiber network. And again, perhaps if we think in the standpoint of definition, you think each neighbor is directly connected to a neighbor on the left side and neighbor on the right side. So if you think about it, it goes in a ring or a circle. The downside with this type of topology is this. If I get a break in the link, anywhere in the link, guess what? The entire communication dies for this topology.
So anytime you get, what I call, the SPF, single point of failure, then of course, Murphy's law says if it's gonna break, it's gonna break. And you're gonna be out. So this is, not redundancy built that in. If you think about solutions that have been proffered sometimes you'll see this, where it's a dual ring. So that if one of the links break, this one's not touching this one. I left it up there. If this one breaks, I still have communication on the other one while I'm troubleshooting why this one broke.
And fixing itself. That is a brief rundown of the ring. Now, moving right along, let's do this. This would be, and of course, you don't think of an actual star, the topology name. In fact, I'll put it up here, star. If I were to actually draw it looking like, of course I'm not particularly gifted artistically in my drawings. So let's see if I can get close.
Notice that all nodes are connected to some central point. And oddly enough, this does look kind of like a star. You know, if I do my little, make it pretty. Guess what? We have a star. But it doesn't have to look like that. All a star topology really means is that all nodes are connected to some central point. Once again, we run into the same problem though. What happens if my central point dies? Then of course, I have no communication. We have lost our ability to communicate.
So sometimes you'll see where they'll have an extended star. Where perhaps you'll have a point over here connected to some nodes, connected like this. So they call that an extended star. And you have similar things out here. Where if something over here dies, these nodes that are not connected to it, they can continue to function and have connectivity. So being an extended star. Now, one other thing to keep in mind about topology.
Think about it from the standpoint of a similar drawing to before. Let's say I have a node here. I have a node here. I have a node here, here, maybe one more over here. These, in this case, I'm representing a wide area network. In other words, this could be Hawaii. This could be Arizona. This could be, I don't know. Let's put Florida over here, Texas over here. I don't know, Michigan, who knows? But these are the spirit geographic locations.
So when you see a diagram like this, typically, you're gonna think of wide area network connectivity. So you want to have, or at least depending on how much money the company's willing to spend. Your connectivity, if the IT department gets to vote, the IT department says, hey, we need everybody connected to everybody for absolute maximum speed. Well, we'll connect everybody up. Did I miss any? Let's see, I got one here. I got one here.
So everybody seems to be connected to everybody, right? I didn't miss any? So here's the question that comes to mind now. If you are designing this network, and you want. And by the way, this is called full mesh, because everybody has a direct connection to everybody. This is also, of course, the most expensive. Because if we put locations on these, you realize that a wide area network type of topology is going to be typically served by a monthly payment made to some provider.
Every single one of these represents some monthly bill that the company has to pay in order to have connectivity. So it's costly when you have full mesh because everybody's talking to everybody. Sometimes companies will say, well, I do need a direct connection here. But if I want to talk here, I don't have to have this direct connection here. I can go this way. So I still have the ability to communicate without spending the money for the full mesh. However, if you are trying to perhaps prep for an exam, one of the things you might want to know is. Say, if I'm told, one, two, three, four, five nodes or five locations, and I want a full mesh solution.
How many total connections do I need? Well, I could take the time to try to count them. Say, well, I got one here, two, three, four. Did I count this one already? You know what? This is not very efficient, prone to error. A much better way is to use math or let math help you. Turns out that there's a formula that if you're solving. You have to know the number of nodes. And you want to know the number of connections? Then the node formula looks like this.
Number of nodes times the number of nodes, minus one, divided by two. Now, what did we say? We have one, two, three, four, five nodes. We plug that into the formula here. That's five, five minus one is four, there's four, over two. 20 divided by two equals 10. So rather than count them, I can determine the number of connections by just plugging the numbers into a very simple mathematical formula. This is just a brief introduction of topologies.
You need to know as a network admin. You need to have a familiar knowledge with various topologies just because they could be any combination of these. And as the enterprise of other networks expand of course, they can have hybrids. Where they use one aspect of this in one area. Another aspect in another. So get familiar with these. Use that formula to help make your life easier instead of counting connections. And you'll become a much more efficient network admin by this baseline understanding of network topology.