Recently, while working for a client that is hosted on Amazon’s EC2 product, I came across a rather odd occurrence, something that generally isn’t supposed to happen, which I captured in this screenshot:

In case you can’t see the image, the problem I ran into was this: “Amazon is currently out of capacity.”

The fact is that any shared system is going to eventually run out of capacity. An elastic rubber band, no matter how big and stretchy, will eventually snap. There are ways to help mitigate that, but at the end of the day, that is a known design decision that allows Amazon EC2 to work a ridiculously high percentage of the time.

A different philosophy is that of our Private Clouds. When you have a Private Cloud, you know precisely where your threshold is for resources; as long as you do not consume all of your resources, you can know—not with a “high percentage”, but with a 100% assurance—that you are not going to run out. And while Amazon’s “out of capacity” error cleared up when I clicked to restart my instances, there’s no guarantee it would recover immediately. Imagine your web site running out of capacity in the middle of the day… but not because of anything you did!

The fact is that it can happen not just to Amazon, but on any shared system—even our own shared hosting or Cloud Application environments! The one way you can be assured to always have a guaranteed amount of always-available resources is by having those resources be private. And one of the only ways you can have the assurance of private resources with the flexibility and power of the cloud is with an Agathon Group Private Cloud.