What is a CDN (Content Delivery Network)?

How much video do you stream every day? How many times do you shop on very large web sites? The amazing thing about the internet is… pretty much everything is instant.

Just five years ago, we all just accepted that there were times when the internet was too busy to stream. Today that’s rarely the case, and when it is, the problem is usually your local neighborhood not the greater internet.

Did you ever wonder how that happens?

Content delivery networks make it happen.

A content delivery network, or CDN, is a group of servers, sometimes spread all over the world. They are linked together and all have the same stuff on them. When someone asks for that content, the closest or most available server is contacted. You get your kitten videos faster, and the whole system is more robust than if it were located in just one place.

CDNs are a part of the backbone of the internet. We very rarely think about the wires, radio transmitters, server farms, IT professionals, and mindboggling number of hard drives it takes to make the internet work. Yet, work it does and it’s pretty amazing that it does.

The idea: avoid a nuclear holocaust

The entire idea of the internet came from research done in the 1960s. Back then, there were just a few hundred computers in the world. The government worried that an attack on a small number of them could knock out a big percentage of our computing power. Since the government wanted to use computers for national defense, this was a problem.

With an “internet,” information goes from computer to computer over the most available path and there are multiple ways to get from computer to computer. This means that just cutting one phone line wasn’t going to take everything down.

CDNs take that same idea but add the idea of distributing content. Because there are now millions (possibly billions) of computers out there now, it’s possible to store the same stuff on more than one computer. CDN servers communicate with each other using the internet’s main lines, and are constantly “replicating” their information between one or more locations. At each location, more than one server may be used so that there is even more redundancy.

Who runs these CDNs?

There are a variety of companies out there who do this sort of thing. Amazon and Akamai are two that have been at it since the 1990s. Akamai might have been the first CDN, because I remember seeing their web address in my browser way back in 1997 or so.

While a lot of companies have sprung up to do this sort of thing, most of them are still reliant on local internet service providers to give “last-mile” service. Cell companies and local land-based internet companies pull the content from CDNs and get it to the home. As I said, that’s usually where the problem ends up being.

Unless, of course, that “last-mile” provider is also a CDN. That’s where I think AT&T has all the others beat. AT&T and Verizon operate their own CDNs, meaning that the stuff that flows to their servers can also flow smoothly to their customers. It’s just another reason why it’s great to be an AT&T customer right now.

If you’re not an AT&T customer yet, fill out the form below. We’ll get that fixed up for ya 馃榾

About the Author

Stuart Sweet
Stuart Sweet is the editor-in-chief of The Solid Signal Blog and a "master plumber" at Signal Group, LLC. He is the author of over 8,000 articles and longform tutorials including many posted here. Reach him by clicking on "Contact the Editor" at the bottom of this page.