No technological concept has elicited more confusion and marketing hype than the "cloud." The confusion starts with the simple fact that the word itself seems intangible and fuzzy, and is further complicated by marketing people who want to attach "cloud" to every new marketing product they have. And to make it even more complicated, before we had the "cloud," the Internet was typically represented iconically in diagrams as a cloud. The latter causes never-ending confusion on the part of end-users, managers and marketers who mistakenly believe that anything related to the Internet is "cloud computing."
Everyone (especially marketing people) has a different definition for the cloud, although the generally accepted NIST definition
is that it is a pool of shared and configurable resources that can be provisioned and scaled on the fly. NIST further divides cloud computing into three service models: software as a service, platform as a service, and infrastructure as a service. And while the delivery mechanism is usually the public Internet, it does not necessarily have to be so.
An evolving model
Like anything technological, cloud computing, even according to the NIST definition, continues to be an evolving paradigm. What it comes down to is simply this: Cloud computing seeks to accomplish what technologists have always tried to accomplish, ever since the US army created ENIAC during World War II, and that's the ability to increase capacity, and to allow more people to do new things they couldn't do before, with fewer resources.
Today, cloud computing has gone beyond the proof-of-concept stage, though it's still early. With vendors and IT users putting more emphasis on many different aspects of cloud computing, more cloud services are emerging, putting us today firmly on the brink of the dotcloud boom—a slightly more sane version of the dotcom boom of the '90s that will change the way we do business, the way we work, and even the way vendors create and sell technology.
A recent survey showed that those who are interested in the cloud actually have a much broader definition than does NIST, and add to that SaaS/PaaS/IaaS triad things like utility computing
, web services, and managed services. And those of you who remember the dotcom boom understand that, despite some people having lost a lot of money towards the end, a lot of very useful technology that is still valid today came out of that period. The same holds true for the dotcloud boom, and we can look forward to new innovations. For example, arguably the next most powerful technology paradigm is mobile computing, and the combination of that and the cloud is a natural—and already starting to take hold in the marketplace with a new "mobile cloud" model. Once again, it's a good time to be a dotcommer.
Is it all just marketing hype?
There's little doubt that there is marketing hype involved. It's the nature of business. But behind the hype, there is, ironically, something definitely solid behind the "cloud" model, and it's one that companies of all sizes will be hard pressed to avoid in the future.