The Physical layer of the OSI model isn't the most exciting or interesting to work with, but it's the foundation for everything we do in networking. The same goes for the physical side of networking - whether it's cable types, network topologies, or network cards, it's not necessarily the most exciting study you'll ever do, but it is the most important. After all, if a network has physical issues such as a bad Network Interface Card or mismatched cables, there's no way the network can work properly!



My next few Network+ exam tutorials will discuss the different network topology types, and we'll start with a look at the dreaded bus topology. After we define it, I'll tell you why I call it "dreaded".

The bus topology is a shared medium in that multiple devices are going to use it to send data. If one host is sending data, none of the other hosts can send data until the sending host is finished. Also, all hosts on the bus will see packets that are destined for one particular host.

A bus segment has to stop somewhere, and the signals transmitted by hosts on this particular bus will be stopped by terminators located at the physical end of the segment.

Now, why did I call this topology "dreaded"? There's more than one reason:

The topology is shared, so only one segment can possibly send data at one time, which is highly inefficient

Bus topologies are not scalable. By "scalable", I mean that we can't add to it in an efficient manner. The more hosts we add to that physical bus, the more hosts we have that have to wait to transmit, the most hosts that have to examine the destination address of packets they won't end up accepting, etc.

Bus topologies are subject to a single point of failure, and we want to avoid that at all costs!

What do I mean by "single point of failure"? If we have nine hosts on a bus, only one can transmit at a time. That's bad enough, but what happens if there's a problem with the physical bus? There's a big problem, because bus topologies are not fault-tolerant.

Then you've got nine workstations that can't send data! The bus segment is a single point of failure - there is no backup way to send data, and an error anywhere on the bus will prevent data transmission by any host. Get used to looking for and preventing single points of failure, because these have to be guarded against in everything from physical network setups to routing protocols, and everything in between!

The only "benefit" to bus topologies is that it uses less cable than the other physical topologies we're going to examine. Cables are pretty cheap, and the drawbacks far outweigh the potential benefits.

I personally recommend you avoid bus topologies in the real world, but you must know all about them to pass the Network+ exam. In my next tutorial, we'll look at star and ring topologies. Until then, keep success as your destination, and keep studying!