This is the time of year when you traditionally review key events of the past year and think about your expectations for the one that is just starting. Two years ago, I reflected on blogging, why I did it and where I thought it was heading. It was all still very new to me - having started this blog several months earlier in May of 2005. No longer a novelty, for me personally or for the world in general, blogging is now a part of the general environment. Blogging has achieved liftoff, that is, it has made the transition from being used primarily by leading-edge adopters to being embraced by a wider marketplace. Technorati is now tracking over 1102 million blogs.
Last year, I wrote about virtual worlds as a key trend that would similarly take off and become more widely accepted in the marketplace as the year progressed. While the level of activity and marketplace experimentation is increasing, we are not quite at the liftoff stage yet. I am convinced that highly visual, collaborative platforms and interfaces - which is what virtual worlds are all about - will become increasingly important to help us deal with IT applications in a much more human and intuitive way. But as is often the case, it is much easier to predict that something will happen than it is to predict when it will happen.
So, for 2008, what do I find as intriguing as blogging two years ago and virtual worlds last year?
I have frankly been struggling with this question. My mind is full of ideas, but none as specific as those of the last two years. I started to wonder if perhaps I am suffering from a bit of writer's block - "a phenomenon involving temporary loss of ability to begin or continue writing, usually due to lack of inspiration or creativity." What a downer that would be to start the New Year!
So I went back and looked at some of the blogs I have written in the last few months, and then it hit me. What I am sensing is that perhaps 2008 will be a key year in the transition to an IT-based knowledge economy. Something this large cannot possibly involve any one trend, which accounts for my struggles. Rather, there will be a set of indicators that something is afoot. One such indicator is the emergence of the kind of global technology platform that is needed to support an information-rich, knowledge-based future.
In the last few months we have seen a number of major stories on Cloud Computing - such as this BusinessWeek story, and this one in the New York Times. Most of these stories focused on Goggle, which runs the biggest cloud-based data center.
What is cloud computing? It is basically an Internet-based network made up of large numbers of servers - mostly based on open standards, modular and inexpensive. Clouds contain vast amounts of information and provide a variety of services to large numbers of people. Users of the cloud only care about the service or information they are accessing - be it from their PCs, mobile devices, or anything else connected to the Internet - not about the underlying details of how the cloud works.
Google's cloud, for example, consists of hundreds of thousands of such commodity servers, supporting search, maps, news, e-mail and many of the other services being offered by Google over the Internet. Yahoo, MSN, Amazon and eBay have their own version of such clouds. In November, IBM announced that it would bring the benefits of cloud computing to enterprise data centers.
Why do I believe this is such a big deal? As the BusinessWeek article says, "a move towards clouds signals a fundamental shift in how we handle information. At the most basic level, it's the computing equivalent of the evolution in electricity a century ago, when farms and businesses shut down their own generators and bought power instead from efficient industrial utilities." Many enterprises will want to evolve and become such information and services utilities to their business customers or to consumers.
And for any enterprise, clouds hold the promise of orders of magnitude more information and computing capacity, to be used in a variety of innovative ways - from providing a diverse set of services to a business’s customers, to the anticipation and avoidance of risks, to more efficient and intelligent ways of managing the business.
How easy is it to build such highly scalable enterprise data centers? Paradoxically, even though the individual components are inexpensive commodity servers, when you aggregate them in such large numbers - hundreds of thousands, going to millions - the overall IT infrastructure has to be very carefully developed and operated. You are dealing with systems at such a scale and complexity, that some of us have been turning to biology as an inspiration for how to deal with such complex systems.
You need a much more disciplined overall data center architecture, with particular attention to:
- integration of all the components, not just with each other but with the other systems in the enterprise data center;
- virtualization – that is, making the overall cloud appear as one virtual system;
- management – a huge challenge, because it is directly related to keeping the system working all the time, regardless of individual component failures, and doing so while keeping down the overall costs of managing the system; and
- efficiency – making sure that the system delivers the most computing power for the least use of energy.
For sure - lots and lots of exciting things to keep us busy in 2008 and beyond.