We often reference our network topologies as standard types, such as star, ring, and mesh. In this video, you’ll learn about physical and logical network documentation using these standard network topologies.
<< Previous Video: Assigning IPv6 Addresses Next: Common Network Types >>
If you’ve been tasked with creating some documentation of your network configuration, you might want to create a physical network map. This would show us exactly what devices might be on the network, and it would show the physical cable connections from one interface to another.
It’s also common for people to create a physical map of where devices might be installed in a particular rack. That would allow you to go into a data center, to a specific rack, and identify exactly the piece of equipment that’s listed in your documentation.
These days, of course, you may be working with network equipment that isn’t physical at all. It may be virtualized. Or you may want to provide a higher level view of how the network connects together.
In those cases, instead of creating a physical network map, you may want to create a logical network map. And regardless of the type of map you plan on creating, you want to use some specialized software to do this.
In Windows, you might want to use Visio. Mac OS has OmniGraffle. And there are also, third-party websites that allow you to create these graphics inside of your browser itself. These would be high level views that would show how a Wide Area Network may be connected. Or it may show the logical communication of an application, as it goes from a web server into middleware, and then, the database back end.
If you’re working with a third-party, it’s common to use this logical network map to show how an application data flow may be working. Or to see exactly how the network may be laid out across a large area. This is a good starting point, and then, you can work down further, to find more details in the configuration as the collaboration goes along.
There are many different ways to represent a network topology in these diagrams. One network topology type might be the star network. This is a very common network type in today’s switch networks, where you have a centralized device– like a switch– and all of the devices are connected directly to that switch. The switch effectively, sits in the middle, and everybody else connects to it over these Ethernet links.
Another common topology type is the ring topology, where all of the devices are connected to each other in a ring form. We used have token ring network as very popular local area networks, but now that those are no longer very common, you would see ring networks and things like Metropolitan Area Networks and Wide Area Networks.
It’s very common to see these ring networks implemented as dual-ring networks. That way if there is a break in the network, it can loop onto itself for fault tolerance and maintain the uptime of the entire ring by providing that loop onto itself.
For even additional fault tolerance, you may want to consider implementing a mesh network. This is one that has many different links to be able to get to the same place. It might have all of the different locations connected to every other location. Or it might be a partial link, whereas some locations are connected to other locations.
You would commonly implement a mesh network if you were looking for something that was redundant and fault tolerant. Or if you needed to balance load between multiple links. You often see mesh being used in Wide Area Networks. Since there are so many opportunities to have a link go down, you would have redundancy and a way to route around the problem using this mesh network topology.
A bus typology is a connection that all other devices connect to. Very common use of bus technology was in early networks, where we would have a single co-ax cable and connect all devices onto that co-ax cable so they could communicate to each other. This was obviously, a very easy network to implement because you simply used a single cable that everyone would connect to, but if you did have a break in that cable, it would disable the entire network.
One place you might find a modern bus technology is in your automobile. This is called a Controller Area Network, or a CAN bus type of network. And that’s where all of the internal devices inside of your car are able to connect to this single bus and communicate between each other.
Although, we don’t use cables to communicate over our wireless networks, there are still some logical network topologies that we use with our 8.02.11 wireless networks. One of these is the infrastructure topology. This is the one that most people are using when they are communicating with an access point. You adjoin an access points network and then, you’re able to communicate with everyone else who happens to be on that network.
But you don’t have to have an access point to be able to communicate with 8.02.11. You could have devices communicating directly with each other, without putting an access point in the middle. We call this ad hoc networking. You don’t need any additional hardware. You simply configure both end stations to communicate to each other over this wireless frequency.
And today’s Internet of Things wireless networks often use mesh networking. These are individual devices that are able to discover each other and use each other as a large cloud of devices. This allows these devices not only to self-form into this mesh cloud, they can also self-heal. If a device is disabled, they can simply use this large number of devices to route around the problem and enable communication using a different path.