Don’t marry your data to the Cloud without a pre-nup
by Craig Jeske, on Mar 5, 2013 2:46:50 PM
Cloudity cloud cloud cloud. This is what I hear in my head nearly every time someone says they “have” to move their data to the Cloud. And every time I hear that I have concerns. Had a great round table discussion last week with a group of executives from various companies around the Denver market put on by GTRI and SympIT (http://www.sympsocial.com.) The topic started out as a wide-open conversation on compliance in the cloud. As with any situation where a group of people sit down and simply start talking, the conversation grew a life of its own, and therein lies the beauty of such an event. There was no company presentation from some sponsoring vendor to put everyone to sleep, it was a discussion around why Cloud and how Cloud. One topic specifically drew my attention as we were talking, which is the age- old question of who owns the data. This question is even more critical when we are putting our data in the Cloud. Who owns my data when I put it in the Cloud? This questions fundamentally feeds an entire string of other issues such as what is the expectation of removal time, retention time, and portability when I put my corporate data into the Cloud? As we were sitting around the table discussing these concerns it became clearer than ever that we as organizations are actually marrying our companies to the Cloud. As a wide-range of public personalities have taught us (Donald Trump, Paul Mcartney, Mel Gibson, Tiger Woods) a marriage without a pre-nuptial agreement can be costly.
An IT exeutive or finance leader may sit there and think, “but it’s my data…of course I own it and can do what I want with it.” And at the end of the day that is a true statement. You as a company most likely do own the actual data when you place it into the Cloud. The question at the time of the breakup is how quickly do you get that data back? And more critically, how easily can you move that data from Cloud Company A to Cloud Company B? What is the migration plan? Say you are hosting your data in Amazon Web Services and you have outgrown their model. It is time to move into a more purpose-built offering from someone like Hughes Cloud Services (HCS), ViaWest, or Ajubeo. Amazon runs a version of Citrix XenServer for their massively scalable cloud offering. HCS, ViaWest, and Ajubeo and many others run VMware-based cloud services. What is Amazon’s motivation to make your transition out of their offering to a competitor? More critically, how long will this take and how painful could the migration be?
This is where a good pre-nuptial agreement is paramount before entering your cloud marriage. To be able to take advantage of the portability of cloud computing you need to be confident you have an exit strategy and an expected response level when you decide to terminate your contract with a particular provider. The type of cost issues a firm could face would include simple downtime, legal fees, overtime, specialized consultants, custom app-dev, and contract extension fees if the fine print is not clearly defined for break-up matters related to data handoff. No matter how much you trust your cloud provider today, unlike a marriage I don’t think any Cloud provider can truly promise a life long dedication to the other half of the partnership.