Flavor Flav once said “don’t believe the hype,” but should we when it comes to cloud computing? In administrative
circles it seems to be the talk of the town, and certainly has benefits, though it’s still developing as a way to store data. Christopher Dorobek of the DorobekINSIDER recently sat down with Jim Sweeney, Chief Technology Officer at GTSI, to speak about cloud storage, it’s applications and whether or not we should believe the hype.
The fact is, cloud storage as a term is being taken advantage of. It seems like everything is being relabeled as a cloud to get customers, when in fact they may just be data centers. Data centers are different in that they are not geographically dispersed like cloud storage. What can be properly termed ‘cloud storage’ are the three types of clouds: platform as a service, software as a service, and the various deployment models including public, private, community, and hybrid. GTSI breaks down these models and figures out which one will work best for the customers – or citizens – an agency is serving.
Cloud storage is very useful, however there are workloads that shouldn’t be uploaded to clouds today and maybe never, such as classified and sensitive data. When processes moved to virtualization, it was a great thing, but it was important not to start out with critical processes – moving to cloud storage should be looked at the same way. Great examples of a “first step” when moving to cloud storage would be e-mail or test and development. With test and development, an agency may need many machines quickly to test new software, but they do not need to actually own the machines – therefore it’s best to use cloud storage.
Security problems are often touted as a reason not to move to cloud storage, but it can also be an excuse. It’s true that classified data probably shouldn’t be moved to the cloud, but in many cases, cloud storage is more secure than today’s data centers.