I’ve been meaning to write about cloud computing for quite a few weeks. As a proponent and advocate for autonomic computing (servers, storage, security, and network), I have a hard time reading most blogs, analyst reports, and articles on the subject. Somehow, cloud computing is being described as the “next generation of grid computing” or “the natural evolution of the Internet”.
When I started my career at AT&T, people would often draw a cloud as the wide-area-network connection between two end-points. Sales reps and even technical sales people would talk about the benefits of the cloud and how it was superior to our competitors. To paraphrase a mentor of mine, “people who talk about clouds and draw clouds on white boards often have no idea what’s in the cloud and how it really works; don’t be one of those people.”
Today’s cloud computing discussion is fraught with old thinking, recycled products, and muffled innovation. Complicating matters is an effort to create a new architecture or a “Universal Cloud Interface” or API to explain cloud computing. Of course, these efforts are sure to benefit the usual suspects. This is analogous to a big oil company discovering a plentiful and easy to obtain alternative source of energy; while it may exist, where’s the profit in bringing it to market until the trillions of dollars of oil is pumped from the ground?
While working at Cassatt, Bill Coleman (CEO) often talked about “an inflection point” whereby the industry would break free of old-line thinking and realize that there is a solution. Are we really at this inflection point? Are we ready to accept that current silos, deployment, and utilization of servers, networking, storage, security, applications, and even virtualization within the datacenter are at a breaking point?
Whether it is cloud computing, utility computing, grid computing, web 2.0, software-as-a-service, application service providers, or whatever you want to call it, the industry must first face the titanic challenges of this revolution prior to realizing the rewards. We must be willing to re-examine the fundaments of the datacenters such as hardware, operating systems, system resources, programming languages, management systems, artificial intelligence, people (may be the most important of all), and more. For the future of the “cloud” is dependent on fundamental changes in all these areas and deep innovation must emerge.
PlatenReport 
Floyd – What do you believe are the fundamental changes that are required? If its not an evolutionary change what is it that you believe will trip the inflection point?
I’m curious because I’m not sure that 95% of enterprises are capable of fundamental change if it occurs at a specific inflection point. Enterprises typically evolve more slowly – closer to geologic time.
That said, the Millennials, the acceptance of technology as part of our daily lives, and agility of more successful businesses are going to put a lot of pressure on change in the enterprise and we will see evolution at a quicker pace. I’m just not sure it will be part of any punctuated equilibrium.
~Ron
Ron – Great Post. External clouds will certainly move faster with “new” applications while internal clouds will move much slower due to “legacy” applications and policies. However, without proving value the internal cloud idea will never take flight. The fundamental changes are centered around management, control planes, application design, and more. Remember when people said the mainframe is dead, how about NT, cobol, etc.