The cloud in the forest

Computing models tend to change gradually over time in the same way that forests do. So it is with cloud computing.

Gordon Haff
Gordon Haff is Red Hat's cloud evangelist although the opinions expressed here are strictly his own. He's focused on enterprise IT, especially cloud computing. However, Gordon writes about a wide range of topics whether they relate to the way too many hours he spends traveling or his longtime interest in photography.
Gordon Haff
3 min read

Irving Wladawsky-Berger, chairman emeritus of the IBM Academy of Technology, offers what I think may be the best way to think about the evolution toward cloud computing.

We should view computing models much more like forests than trees.  These computing model forests have a variety of different trees, and the transition between them is gradual, not abrupt. With the passage of time, as you walk around them you begin to see new trees, but the old ones are still around. But one day you realize that the forest you are now walking through is markedly different from the one you were in twenty years ago.

Put another way, it's a piecemeal evolution--neither a new term applied to the same old thing nor an overnight sea change in the way that all computing is done. Terminological debates will doubtless continue. And there are plenty of unanswered questions about exactly which new "trees" will thrive and which will die out. However, we're starting to see come consensus emerge around at least some broad outlines of the cloud's evolution.

As Irving wrote after the recent MIT Sloan CIO Symposium:

A lot of the benefits of cloud computing, such as virtualization, shared infrastructures, highly disciplined systems management, flexible deployment and scalability are of value to just about all data centers and service providers, whether you run them as a private clouds providing services to only members of the company, or as public clouds open to everyone. There is also general agreement that you should make cloud deployment decisions on a case-by-case basis, especially decisions as to which applications should be run on private versus public clouds. Public cloud deployments make the most sense for highly commoditized, standardized, mass customized applications.

There was also a broad consensus that infrastructure savings and flexible scaling where key adoption drivers for cloud computing. One panelist remarked that "The benefits of cloud computing start and end with the dollars you can save.  But it can also help getting your best people away from working in areas that can be outsourced to the cloud. This way you can allow your best people to focus on an area that drives differentiation for your company."

There are a few points that I'd like to highlight:

Cloud-like computing architectures within organizations are garnering a lot of interest. Whether you call them "private clouds" or just the next iteration of service-oriented architectures (SOA), the bottom line is that organizations--especially larger ones--are far more interested in leveraging the approaches embodied by cloud computing than they are in actually hosting their applications elsewhere.

There's also interest in the public cloud, but primarily for standard off-the-shelf applications. This is, in fact, wholly consistent with the pattern of IT outsourcing going back years. Payroll was perhaps the first application to be farmed out in a widespread way. Important? Sure. But also largely standardized and in no way a source of competitive advantage.

Distractions matter. Enterprises can arguably do a lot of things as cheaply in-house as can a third-party. However, do too many non-strategic things and you have the corporate version of too much "stuff" rattling around and getting in the way of what's really important.

It's arguably been the dramatic "Big Switch" take on cloud computing that has led to so much of the hype surrounding it. But it's arguably the enabling technologies--as adopted in thousands of distributed organizations--that are more relevant for any reasonable planning horizon.