One of my clients is a large US Government agency. As with all agencies, the Government's Cloud First initiative is pushing them to move to the Cloud. The reasoning is solid:
- lower the cost of IT
- streamline IT purchasing
- leverage the elasticity and self-service benefits of the Cloud
But this agency has found themselves in a tangle, due to the complex and unique IT challenges they face. They support dozens of critical systems, many of which thousands of lives depend on. So as they go through the list of existing applications and systems, they're basically making a thumbs up/thumbs down on whether to move each app to the Cloud. Unfortunately, all too often the thumb ends up pointing downward.
Sure, there are security concerns. Sure, latency is an issue for many of the apps. Sure, there are any number of other reasons to nix the Cloud for this app or that. But by considering the Cloud to be a single thing deserving of a simple binary yes or no, they are doing the Cloud -- and themselves -- a disservice.
It would have been better had they not heard the term Cloud at all. If the nebulous marketing terminology fairy had never waved her magic wand, we wouldn't be talking about Cloud. Instead, we'd be talking about a shopping list of various capabilities, solutions, and service models that we could choose from.
Third-party multitenant SaaS. Single tenant hosted SaaS. Public PaaS. On-premise PaaS. Virtualized servers. Virtualized servers with automated configuration. Virtualized storage. The list goes on and on.
Because this agency is understandably risk adverse, if there is a risk to one aspect of Cloud, that risk may be sufficient to nix any aspect of Cloud, even when the risk doesn't apply. By buying into the sometimes useful fiction that Cloud is only one thing, this agency is frittering away tax dollars that might otherwise go to providing better services to citizens.