Sun shining through the clouds

After year of disillusionment, agencies should get clear on cloud realities

If 2012 was the year that many government agencies began moving in earnest to the cloud – and one assumes the benefits of doing so – 2013 promises to be what might be called the year of cloud disillusionment.

While agencies will continue their transitions to the cloud, they are likely to find out that the innovation, cost savings and efficiencies it brings are more complex than they have been led to believe. The good news is that this period of disillusionment should be short-lived and will lead to a scenario in which agencies emerge with a better understanding of cloud computing and its benefits.

Let’s look at the evolution of this story. Cloud was introduced as an innovation that people didn’t really understand, so a couple of years ago the National Institute of Standards and Technology went to great lengths to help us define it. It developed good documentation, architectures, terminologies and taxonomies to help us begin intelligent discussions about what was a new concept to everybody.

The problem was that cloud became an all-encompassing term. So NIST broke it out as separate IT deliverables: infrastructure-as-a-service, platform-as-a-service and software-as-a-service. Fast-forward 18 months, and pretty much every vendor now claims to be a cloud provider, and almost everything becomes cloud. A network engineer may think of cloud as an infrastructure solution. If I talk to a software developer, he’s thinking platforms. And if I talk to an end-user, he’s thinking of software. So the very word tends to lead to confusion.

We are now at the point where even custom applications – applications that aren’t widely used and that don’t require the scale associated with the cloud – are being purchased as software-as-a-service or as cloud solutions. That really distorts the definition of what made the cloud innovative and unique. An application that is unique or customized to an organization can be bought as a service but it can’t really be considered a cloud solution in terms of its economic model where many people would be reusing the same software, infrastructure and platforms. The term cloud doesn’t apply, because the application is so unique. And a lot of what the government does is very unique, from issuing passports to tracking the various elements for national security purposes.

One might think e-mail, for instance, which is very similar in most agencies, would be easy to transition to the cloud. But even that can require the use of multiple cloud solutions and can require a lot of work to integrate services and security around it all.

To further complicate matters, according to some estimates, 80 percent or more of the government’s workload still resides in legacy systems. It’s going to take a lot of new ideas and sophistication to transition these workloads to take advantage of cloud. I don’t think people realize yet how challenging that is. And as much as people in the industry say they want things to change, they are having a hard time adapting because those innovations cannibalize work streams that have been there for many years.

The way to overcome cloud disillusionment is to become more sophisticated in the way we evaluate the architectural opportunities for changing how legacy systems are managed and operated. An initial approach would be through intelligent application alignment services that look at your portfolio of applications and help decide which ones would or would not be appropriate to move into a shared services environment.

In addition, a cloud broker model, in which a third party is brought in as an agent to optimize cloud-based purchasing and usage, can provide the necessary expertise to take advantage of various components of the cloud and to keep it secure and compliant. 

The biggest savings are going to come when agencies get past the low-hanging fruit and begin to tackle platform-as-a-service. That’s going to be harder, because it will require some changes to the existing application base. And that’s where the pending federal budget cuts come in. The pressure of those cuts is what’s going to enable agencies to make some of these dramatic decisions. It should be an interesting year.

About the Author

Peter Gallagher is group vice president for portfolio solutions at Unisys Federal Systems.

Reader Comments

Mon, Feb 11, 2013 OldCIO

Eventually, folks will get the message that often going to the cloud is more expensive and less secure. We need to do our own research and stop listening to vendors and vendor supported reporters. We have changed our statements of "this will not work for the enterprise" to "this is very complex". In many situations, that is a mistake.

Please post your comments here. Comments are moderated, so they may not appear immediately after submitting. We will not post comments that we consider abusive or off-topic.

Please type the letters/numbers you see above