Cloud computing is at its peak of hype. In fact, it's probably about to jump the shark. Savvy enterprise IT managers know there's little reason to embrace cloud this early in its development.
But, in fact, you should jump into some aspect of cloud computing right away. Yes, cloud is sketching the curve described by Gartner's July 2008 report "Hype Cycle for Emerging Technologies." Gartner analysts say cloud will begin delivering products and services that lead to mainstream adoption in two to five years. By my reckoning, cloud's "getting real" time will occur in 2010.
So, why would you want to dive in now, just as the technology is about to crest that first big roller-coaster peak of excitement and head for its deepest dip -- the one that coincides with us all being sick of it? (Remember when Web 2.0 was a swear word?) Because cloud will emerge from the murky hype cycle, and if you haven't experimented with Internet-based enterprise applications and IT services by then, you may fall behind the curve.
The industry has been heading toward cloud-based computing for a long time. When the change comes, it will be transformational.
Cloud computing embodies two aspects: IT services and software as a service. Cloud began as similar concepts described with different terminology. So, grid and utility computing have become on-demand computing, which includes a raft of IT services. And Web services, ASPs and MSPs have become SaaS and platform as a service.
Whatever the terminology, SaaS is beginning to make serious inroads into the enterprise applications marketplace, and on-demand IT services are no longer a pipe dream.
One reason cloud computing is a hot topic is that things have come to pass that make old ideas more feasible, such as soaring server capacity, more modular Web-based architecture, IT spending's slowdown and the beginnings of enterprise interest in cloud- and SaaS-based computing.
Key questions remain about where cloud sits on the track between marketing and reality: Will it allow the replacement of large swaths of IT organizations, as some predict? Can it avert the need for large outlays for hardware and facilities? Is it both reliable and cost-effective? Is it truly available? Might you find yourself sliding down a slippery slope of rising expenses and customer service woes? What about security and privacy concerns if you outsource management of your data?
For most CIOs and IT organizations, cloud computing is a minefield right now. Its potential to cut costs on new efforts is tantalizing, but it comes with an attendant loss of control that's anathema to many IT pros. We're hearing about companies that have taken the plunge, but I don't think those early adopters are representative of the overall market. Most IT organizations just aren't ready to take a big chance on a technology that's barely out of the gate.
That's probably especially true of on-demand services from companies that want their cloud-delivered storage and virtual servers to become adjunct chunks of your data center. On the SaaS side, many products and services are ready to work, but they're not fully mature. The best is yet to come.
As ethereal as cloud computing seems right now, there is substance there, including real utility and potential cost savings. But to gain any advantage, you have to get your feet wet, build vendor relationships and begin to understand what today's technology and service levels can and can't deliver.
Pick a small greenfield project to try it out. Or weave cloud into a project in a way that extends your application instead of replacing it. Experiment now so you'll be smarter about implementing cloud when vendors have more compelling offers. The ingredient that will drive cloud's eventual adoption won't be technology; it'll be cost savings and ROI. Don't wait for the pressure and allure of its potential cost savings to build before you get up to speed.
Scot Finnie is Computerworld's editor in chief. You can contact him at scot_finnie@computerworld.com.