I've been tracking some industry offerings around these "cloud-in-a-box" packages that claim to provide plug-and-play cloud computing (such as this recent announcement). They offer limited combinations of hardware (servers, switches, network fabric, and storage) and software (virtualization, OS, and importantly... management tools), and they come pre-assembled, each in a shiny package with blinking lights and a red bow. And if you run out of cloud, buy another box and park it next to your first, and then you'll have twice the cloud. What's not to like?

At first glance, this idea would appeal to any deep-pocketed enterprise or service provider that is seeking to realize the benefits of cloud computing while maintaining direct control over the physical infrastructure (what's being called a private cloud). Certainly, the "pre-assembled" nature of cloud-in-a-box makes things attractive (Spoiler Alert: They don't come in one box like the one you saw on the tradeshow floor. They come in many pieces as if the parts of a data center showed up on a truck.)

And, from a vendor standpoint, offering these pre-configured clouds, in several different sizes, simplifies the sales, support, engineering, and fulfillment processes by reducing the number of potential configurations that are available for purchase.

But I also see some issues with this model that make me wonder if the cloud-in-a-box approach is going to take off. First, these things are expensive. Justifying the cost will be challenging for anyone but the largest telcos and service providers. And related to this, IT has multiple buying centers. There is the server guy, the network guy, the storage guy, the virtualization guy, the application guy, etc. How often do these people get in the same room and say "Hey, we all just ran out of resources at the same time. Let's all get together, pool our budgets and buy something that will collectively meet all our needs!"

And beyond the issue navigating corporate politics to agree on buying one of these, the bigger question is "Are they the right solution for the customers - and the vendors who are trying to sell them?" Per usual, I have doubts.

Customers know what they want. Or at least they think they do. And every customer has different priorities and criteria. Well, with this cloud-in-a-box model, customers get very few choices and vendors will be unwilling, or unequipped to handle alterations. The management software these boxes use to provide cloud-like services, are only built to run on the short list of pre-set configurations and alterations have not been tested. Making an alteration would reduce your cloud in-a-box to a very expensive pile of hardware.

Likewise, from the vendor point of view, customers will always want a customized solution. Will the vendor sales people learn to say "No changes!" to their high profile customers? Of course not, they'll run to the engineering team and say "What would it take for us to make our cloud-in-a-box work with alteration A, B, and C. This could be a BIG win for us! All hands on deck!" And there goes the roadmap. Sound familiar?

Finally, let's say a customer rallies the troops and goes ahead and buys one of these and it works. One of the known attributes of cloud is elasticity. So yes, they could buy another box if they need more capacity. But if the customer wants to manage their greater pool of resources, including the rest of their IT infrastructure alongside public cloud resources, will the cloud-in-a-box play along? Today, I don't think so. As time goes on... maybe? But will the vendor and customer business challenges of clouds-in-boxes crush the idea before all that integration work happens? I'd love to hear your thoughts.

*Image used under CCL, courtesy of Sheldon Wood.