Two of the most common questions I'm asked are: Aren’t all Content Delivery Network providers the same? And isn’t web site acceleration easy so all caching CDNs are the same?

Those two statements couldn’t be further from the truth.

Web site acceleration is a methodology for having content live on the edge of the network, with delivery to end users from that edge. CDNs offering web site acceleration have to be prepared to constantly deliver millions of web objects. This type of demand is extremely taxing and very difficult not only to deliver but also to report on, making it the most difficult service to deliver - quite the opposite of what most customers believe.

Do all CDNs deploy the same basic architecture for meeting these immense traffic requirements? Not really - different web site acceleration architectures have different edges of the network, with important content and performance implications. One early approaches, which some CDNs still utilize, deployed the “many thousands of caches” method. This method does a good job delivering smaller objects, but may be getting a bit obsolete as the amount and type of web site content continues to grow - individually and in the aggregate. A newer CDN architecture, the “caches at the aggregation points” method, deploys fewer caches that are not as close to the edge - instead they're located at the major Internet aggregation points. This newer architecture generally does better at delivering the larger web objects or videos. You can see these performance differences by monitoring the cache hit and cache miss ratio of a CDN.

So web acceleration architecture is one major difference between CDN’s.

In Part 2 we'll talk about the critical questions you need to ask to determine how well a Content Delivery Network offering web site acceleration will meet your needs.