When EMC decided to shut off its ATMOS Online product offering, it surprised many of us who work in the world of Cloud computing. (See July 04, 2010, EMC Suddenly Cedes the Clouds, Blog by Nicole Hemsoth) Personally, I respect EMC’s decision - competing with one’s business partners is always a delicate matter and often carries more risk than reward. I was, however, taken aback by its abruptness, and the way in which EMC communicated the end of ATMOS to its user base. The service was pulled from general availability, seemingly overnight.

I felt the impact of this decision first hand. I was prepping for an infrastructure control demo that just happened to be integrated with EMC’s ATMOS Online Compute offering. While logged into our control panel, I selected a hosted VM to control, and immediately received an error message. After a “bit” of digging the operations team informed me that when they logged directly into ATMOS they discovered that all our VM’s were gone. Deleted forever. We had been given some warning by EMC, so there was no critical data lost – nor were there any systems that we could not rebuild on another Cloud. It was merely an inconvenience. 

This experience did, however, make me wonder about what would undoubtedly be a real dilemma. 

What would happen if your Cloud provider went dark? 

Whether the “darkness” occurred accidentally or intentionally, my question begs another. What would happen to your business if your Cloud hosted solution suddenly became unavailable or worse yet, deleted and unrecoverable? The answer, whatever it is, won’t be a good one – especially if you rely on the Cloud to house your main production systems.

Nothing punishes a business more than unplanned system outages that lead to employees staring at their screens because email is down…or to massive delays in production lines…or to loss of revenue and revenue opportunity.

Still, fear of a Cloud Provider going dark isn’t going to change the massive migration to cloud services that is happening worldwide. Frankly, I believe that it will be soon become increasingly difficult for IT personnel to build a financial business case to support onsite systems. The ROI just won’t be there when compared to the economies of scale that can be leveraged with Cloud infrastructure.

But what about my question? What do you do if your Cloud Provider goes dark?

The answer might come from looking at what large enterprises often do with their outsourced network infrastructure – they build in redundancy. If one network service goes down, or even if its primary network provider’s network crashes, they often have another “on-demand” network that ensures their business continues to operate in the event of a network failure.

Cloud Services users should adopt a similar strategy to ensure that critical resources are available at all times, and they can do so from within the Cloud. Move to the Cloud with confidence by building continuous availability into your total solution, perhaps even creating a replicated environment that is enabled through a secondary Cloud provider (or through a second cloud infrastructure offered by your primary provider).

Follow this approach and you’ll always feel like a kid with a night-light...never again afraid of the dark.