During last week’s Ingram Micro VentureTech conference in Chicago, there was lots of talk about cloud computing from nearly every vendor present, as well as detailed stats from Tiffany @ the Gartner Group, followed up by 50+ CEOs in a roundtable. Even upon arriving at the airport in Chicago, nearly every column on the way to ground transportation, as well as the doors that exited to the street level ‘arrivals’ area were logo’d with Microsoft’s new Cloud initiatives.

The cloud is having a transforming change on the way IT is delivered, to be sure. Today, I read an article titled ‘Cloud Bubble Warning’ from an editor I know pretty well and see at a lot of our events – Joe Panettieri @ Nine Lives Media. He was making a great point that anytime there’s a major technology bandwagon, there are a lot of companies and people who jump in – and not all of them are going to last. It got me thinking about some of the similarities – and the differences, which can affect us as MSPs, and the clients we serve…

When the .COM bubble burst, a lot of capital was spent on companies that didn’t really deliver a viable product. Thus, the direct effect on the end users wasn’t as significant in most cases when the .COM busted.

But what can happen when the investment capital used to buy the racks, storage, and infrastructure in a data center cage runs out? When funds dry up, and the bubble bursts, the engineers dry up, the assets are sold, and the fragments of data that belong to clients can cease to function. Then the SLAs (however good or bad) can fail too. As we perform ‘due diligence’ on many cloud providers, we find that their applications do not contain the simple APIs to allow the clients to save the data to another source, meaning ‘easy in, almost-impossible out’!

With Internet bandwidth increasing, it’s easy to get data into the cloud, and there is great ‘stickiness’ for the up and coming cloud provider, but it can be potentially devastating for the client/MSP/VAR of the popped bubble if they aren’t left with a way of retrieving the data that runs their business.

I’m a major cloud proponent too, investing extensive capital in our own data center/ cloud infrastructure in behalf of our clients @ AllConnected, and back to the channel @ XiloCore. I’m vested in the growth of cloud computing.

Like a business associate of mine says, the people of Manhattan are extremely dependent upon food service. If just one day goes by without food delivery into the city, the hints of riots can start to take root. The food delivery system is absolutely critical and cannot be interrupted. Like Manhattan depends on food service, users already are and will continue to depend on the cloud as a critical non-interruptable service.

More and more clients are coming to us saying ‘I found a way to move everything to the cloud’. I think it prudent for VARs and MSPs to intercede and ask clients about their ‘B’ Cloud plan, ie, what if the cloud provider isn’t there? A cloud provider (with seemingly unlimited capital) can build a redundant infrastructure that an end user never could. But their SLA usually limits liability to one month of data loss.

The moral of the story, of course, is that if you’re partial to your data, even if it’s in the cloud, back it up with a completely separate system (or cloud).