Cloud-computing predictions for 2011

2011 will be the Year of the X. Next year, Technology Y will kill Technology Z. Something will "die."

These sorts of predictions are commonplace as we approach the end of the year. They have a satisfying finality to them. They're dramatic. They're also, with few exceptions, rarely correct--certainly not in any literal sense.

That's because IT rarely advances in a way that invokes mass extinctions and spontaneous generation. Rather it's a more evolutionary process. There's lots of change but even when rapid the new stuff often doesn't displace the old--and overnight replacements are rare indeed. For example, proclamations about the death of Bluetooth were wildly premature even though that technology didn't live up to early promises.

This is especially true of cloud computing, given that it refers as much to the way the industry is moving to implement IT as the technology it uses to do so. Will those changes lead to broad shifts in where and how computing is done? Certainly, that's what makes cloud computing of so much interest after all. But we're mostly talking about transitions rather than sharp inflection points.

Within that context though, cloud computing is a rapidly developing set of trends that's generating lots of interest and discussion. And those discussions suggest to me some things that are going to be qualitatively different next year compared to this past one and some that will remain elusive.

Less focus on definitions (and dare we say hype?). If we were to do a survey of presentations, articles, and analyst research reports throughout this past year, we'd find that many of them spent a lot of time defining and categorizing cloud computing. I myself wrote a Cloud 101 white paper earlier this year. This sort of content may be old hat to the analysts and vendors who have been writing about or implementing cloud strategies over the past few years. But, as I've discovered to only partial surprise, themes that some of us consider well-worn are still fresh to many mainstream audiences. That said, in 2011, we can collectively start to move on from talking about the big picture while remembering that the future doesn't happen everywhere at the same time.

Getting down to work. Cloud computing in 2011 will be, in a sense, more pragmatic. It will often be more about incrementally leveraging virtualized infrastructure to further automate the management and deployment of applications and workloads. New products that build on top of virtualization will doubtless come to market. However, a lot of organizational energy in 2011 will also go into getting business processes and IT operational procedures in line with the more integrated and autonomic approach to allocating IT resources that cloud computing implies.

"Standards" still won't be. Even in the best case, it's not realistic to expect that all clouds will interoperate perfectly. Proprietary application programming interfaces are often cloaked with a "de facto standard" rubric. But even when standards bodies are involved, cloud standards is still a developing area where one size doesn't fit all. Not all clouds have the same purposes and goals. One might expose lots of options; another might choose to keep things simple. One is most concerned with providing customers with tight controls over service levels; another just concentrates on cost. Regulations in a specific industry can mandate interfaces that relate to audit and compliance that aren't needed by most users. As a result, approaches that maintain portability and interoperability among multiple clouds is going to remain a strategic necessity for the foreseeable future,

Acquisitions of start-ups will continue apace. This one admittedly falls into the "well, duh" category but it's worth highlighting because 2010 has been such an active year, as Andre Yee correctly predicted last year it would be. Cloud computing, to an even greater degree than the server virtualization on which it builds, involves bringing a portfolio of products and solutions to bear on business problems. I'd argue that portfolio is best represented by a loose integration of a lead vendor's products and those of its partners, rather than a monolithic stack. However, it's inevitable that many start-ups end up not having the resources to bring a complete product to market on their own. Or market reach, and therefore value of a given technology may be greater delivered as part of a broader portfolio.

More nuance in security discussions. "Security" often gets used as a sort of catch-all for all manner of concerns that relate to risk in some way, shape, or form. Some of these concerns do in fact relate directly to security--for example, the need for additional layers of protection when multiple organizations are sharing the same physical infrastructure. However, many more "security" concerns are really more about compliance with government regulations and corporate standards, the ability to guarantee service levels, and generally having visibility into how applications run and data is secured. We'll increasingly see cloud-computing discussions move on from generic concerns about risk, privacy, or security to more focused examinations of specific requirements.

Cloud computing brings together trends in both technology and IT operations. It's not something wholesale new but, rather, builds on things like virtualization, automated provisioning tools, and application delivery over public networks. In fact, that it's evolutionary in important respects is precisely an important reason why cloud computing is so interesting. And that evolution will be in full force in 2011.