Blog post

Experts Define Cloud Computing: Can we get a Little Definition in our definitions?

By Daryl Plummer | January 27, 2009 | 8 Comments

A recent sys-con article on Cloud Computing purports to present Twenty-One experts defining Cloud Computing. After reading it I was impressed by the list of experts but fought long and hard to find any actual definitions in the piece. To me, one of the biggest problems we have in IT is the vagueness and lack of precision in all of our work around these complex topics. And this article serves to cement my concern as even the experts use less than precise descriptions to get across their points.

No, in the interest of full disclosure, I was a bit taken aback that Gartner was not listed as one of the experts, given we have one of the most widely quoted definitions of Cloud computing (in fact, it is referenced in one of the expert’s Definitions but taken out of context) in the industry. In fact, I was a primary author so my view point is biased. But, I am trying to put that aside to get to the real issue that we face. That is – how can we expect IT people to be able to strategize and decide on IT direction and tactics if we can’t even describe for them what the real issues are in any consistent way. For that, we need a commonly accepted definition, even if it is not great.

The Gartner definition (modified slightly from our original) “A style of Computing where scalable and elastic IT capabilities are provided as a service to multiple customers using Internet technologies” can be teased apart to get at most of the key issues in Cloud Computing. But if people differ from that definition, I would love to see more definitions that can be teased apart from a consistent foundation as a good place to start.

Certainly, there are many ways to look at cloud computing but the benefits need to be qualified in order to be quantified. Let’s at least ask the experts to start their definitions with actual definitions.

The Gartner Blog Network provides an opportunity for Gartner analysts to test ideas and move research forward. Because the content posted by Gartner analysts on this site does not undergo our standard editorial review, all comments or opinions expressed hereunder are those of the individual contributors and do not represent the views of Gartner, Inc. or its management.

Comments are closed


  • Dan Sholler says:

    Twenty-One Experts Define Cloud Computing | Web 2.0 Journal

    This article is typical of a lot of the discussion around cloud computing.  Many people are trying to come up with “the” correct definition of it. Some of those definitions get into very specific details about what capabilities are needed and how they are delivered. At the end of the day, most of them are quite wrong (this goes to show that you need to choose your experts carefully). While we at Gartner have probably added to the mix with our own definition  (Gartner subscription required. )

    I personally think of this as something much simpler.. this analogy was provided by my colleague David Smith, who has led much of research on cloud computing at Gartner.

    The whole discussion around cloud reminded both of us of the discussion nearly 2 decades ago around client server. At that time, there was an acknowledgment that client/server computing was the model to follow in the future, and a great deal of effort was put into trying to define it. Debates raged over whether the server database had to be relational, whether the client could be a minicomputer rather than a PC, etc. etc. etc. The challenge at the time was that while the principles of client server were known, the best practices and patterns of implementation were still in flux. This is precisely the same situation we are in with cloud computing. This means that whenever we try to define it, we often have to give lists of various mechanisms, practices, business models, etc. (with lots of qualifying weasel words) to make the definition accurate. These ambiguities make the definitions of limited use, since they attempt to encompass a range of practices that are not all, by definition, optimal. Unfortunately we (meaning the IT industry as a whole) does not know which of these will become optimal, and which will be commonplace, and by the time we do know, we probably will not need the definition any more.

    David Smith’s great observation was that in the period of definitional turmoil around client/server.. if you went to an IT professional, and he or she told you that they wanted “a client/server solution”.. there was no ambiguity about what they meant: they meant that they wanted something that did not run on the mainframe. All the other stuff was just “how to” detail. The core concept was that they did not want to put the application on their mainframe.

    The same is true today with cloud computing. When an IT professional says that they are interested in a “cloud computing” or “cloud based” solution, what they mean is that they do not want to own the assets and run the thing in their data center. This “not in my data center” definition, if interpreted strictly, is not correct, but it captures the essence of what is desired. Of all the definitions and discussions referenced in the article, I think Irving Wadlawsky-Berger came the closest in talking about people wanting to buy services or outcomes, instead of computing resources.

    In the end, I do believe that the Gartner definition of cloud computing:

    a style of computing where massively scalable IT-enabled capabilities are delivered ‘as a service’ to external customers using Internet technologies.

    is likely to prevail. (Of course, I am biased in this..) but it is simple, and encompasses the fundamental premise, which is that we are buying services (outcomes) not assets, and it is the internet (and specifically the web) that is enabling this. It also include the notion of scale, which is a critical (and often overlooked) aspect of cloud. Basically, it is the economies of scale that make cloud so potentially disruptive (in the Clay Christensen sense of the word) , and any model that does not have those massive economies (IMHO) is not within the spirit of cloud, since it does not provide significantly different economics. This means that cloud services are by definition “off the peg”. This also means that there will never be a situation in which all services have the cloud-scale economics. For more stuff on scale, see the note that my colleague Donna Scott and I wrote on cloud scalability  (Gartner subscription required)

    Bottom line is that you need to be wary of experts and their definitions and choose them carefully. The real challenge is that while the concept is clear, there is a great deal of variation in practice, and that variation leads to different outcomes. When looking at cloud computing, you must decide what your priorities are, and focus on those practices that have been shown to work in situations similar to yours.

  • Hi Daryl,

    I’m going to go out on a limb and say we’ll get a definition for Cloud Computing about as soon as we get one that makes sense for SOA. 😉

    But seriously, here’s the last definition you’ll ever need.

    Or to be more succinct:

    Shared, scalable, on-demand resources + dynamic infrastructure = cloud computing.

    Thanks for bringing up the topic, I’m only sad to say I don’t think it’s the last time you’ll have reason to question whether we have an acceptable, industry-wide definition.


  • Reuven Cohen says:

    How about just > internet computing.

  • Mark Raskino says:

    Bravo Daryl – forget Cloud for a moment – well done for sticking to the principle that definitions *matter*. It is a constant irritation in this industry that when somebody suggests a term needs clarity of meaning, many will acuse them of indulging in ivory tower semanticism and lacking business focus. In fact the exact opposite is true. Big new IT terms without definition are a FUD-spreader’s dream.
    Words matter. If we have different ideas in our minds when we use a term – it spreads confusion and misunderstanding. Ultimately that leads to inappropriate business decisions and investment failure.
    On the term itself. I still think definitions are missing some key elements. For example ours permits the idea of a ‘private cloud’ – which feels almost like an oxymoron to me. But definitions can only evolve and be perfected though collaboration and debate. So forward … to version 1.1 perhaps.

  • Daryl Plummer says:

    Excellent, Mark, You are correct about meanings and definitions. Too often, I run into the “That’s just semantics” comment from technology people. This comment is in itself oxymoronic in that semantics generally means “meaning” and if meaning is not what we intend to communicate then what are we really doing here? I also find that that phrase is just an excuse not to get real understanding in place or to just get through an agenda. As for the cloud definition, it is far from perfect, but at least with it, we can examine if private clouds actually exist. More to come in research on this definition subject and I will blog the basics.

  • Daryl Plummer says:

    Reuven. Good try? >Internet is convenient and may actually be true. But in the early stages we need more specificity, I think. But then, you have told me that this can lead to more confusion instead of less. I keep that thought in mind every time I go through this.

  • Daryl Plummer says:

    Thanks, Lori. I read those with great pleasure.

  • Tony Antony says:

    I see the challenge and still trying to figure out the definition and meaning of SOA, Grid Computing, Utility Computing..Etc. Recently, I had a conversation with a Sr. manager of a large Service Provider and according to him Cloud is just “Service delivery Platform” – i.e a Service oriented Architecture on “elastic IT” infrastructure ( according to him the “elastic IT” is Utility Computing that is powered by Grid). He was also excited at the business model where “Traditional vendors” has to share revenue with Service Providers to service a customer – everyone wins in this game he added (you have to consider per application basis).

    Back to my point, defining Cloud may be a laborious task – as this means different thing to different people.