In the fight over private versus public clouds, we’re all arguing about the same thing. But we keep doing it because we somehow believe we can confuse the customer into buying “more of mine” and “less of hers” if we explain how our vision of cloud is better. I’ve written about what cloud is or isn’t several times in the past, but, I feel that I missed some important context. The context is what leads me to consider the idea of three different clouds.
So what are they?
- Public cloud – The only “real” cloud as someone like Werner Vogel of Amazon would say
- Private cloud – An on-site or hosted private cloud-capable environment
- Actual cloud -The set of strategies, processes, people and technologies that enable business agility, improved resource management, and faster time to market (among other things). The actual cloud is the real world amalgamation that users end up with and may consist of both or one of the above.
As the idea for this blog was first racing through my mind thought getting feedback on the term cloud from some of my friends would be a great way to further illuminate the disparity of thought on the subject of cloud. As you can see from the tweets and quotes below, each has their own take on the idea of “what is cloud, really?”
Joe Weinman, a SVP at Telx and author of the just-released Cloudonomics defines “cloud” via a retronym: C.L.O.U.D.: a Common, Location-independent, Online, Utility, on-Demand service. He argues that each of these properties generates statistically quantifiable economic value; for example, common resources lead to increased utilization when demands are independent. Such approaches, he argues, provide a sounder basis for cloud decision-making than hand waving.
Others were less scientific, such as Simon Wardley a lead researcher at the Leading Edge Forum, who tweeted: “IMHO – Cloud is a muppet marketing term used to window dress computer utilities as something else so other muppets pay more”
False arguments on both sides of what cloud is or isn’t
The cloud industry benefits from you as the customer being confused by the noise. Each side of the cloud debate is trying to work on your fears that you might be buying into something that’s not secure, will cost too much, or will lock you in etc. This doesn’t mean that there isn’t real cloudwashing going on, but this is where as the customer you need to be clear about what the objectives are. If what you’re buying meets or exceeds them, then it’s good enough. So you can safely ignore the following false arguments:
- Private cloud isn’t real because it can’t possibly be as cost-effective as a public utility type offering such as those from Rackspace, HP, Microsoft, Amazon, Profitbricks, Zumaysys, etc.
- Public cloud is not secure enough
- Private cloud will never scale infinitely and therefore isn’t really cloud
- Public cloud is more expensive
- Private cloud is just a stack of stuff being sold to you by the big vendors
- Public cloud isn’t dependable
- Private cloud software isn’t mature enough
- Public cloud will lock you in to a provider
- Private cloud is just more hardware and software lock in
While all of the above statements are “false” in the right context, they can be true in others. The sad truth is that it’s definitely a caveat emptor market for the cloud service/solution buyer.
Why the what doesn’t matter
When someone aks, “What cloud are you using?”, “Who cares,” should be the answer. What’s important is what you’re doing, not what you used to do it. We’ve been subjected to alternate technologies at every layer of the IT stack for the last 30 years, yet strangely we’re not all on one OS, one storage type, one middleware solution, or one database. As I’ve stated before, IT by its very nature provides us with a painter’s palette of opportunities. What’s important is how you utilize that pallet, not whether you use a specific type of paint.
The correct question should be how are you creating opportunity, or how are you enabling agility. In general the least important addition to your operational model should be the technologies you’ve added. If you’ve added the technology pieces, but haven’t changed your operation model and realigned your business and IT staff, then the tech just doesn’t matter. If you built a cloud based on Captain Crunch and bailing wire, then great, you’ve got a cloud, as long as it’s solving the problems you needed solved and enabling the agility and opportunity you expected. The why of the activity should always be the most important point, because it will drive the how.
Chose the actual cloud
When you’ve created a cloud oriented organizational model, then the technology that supports it is but an enabler. If you can solve the problem most effectively by cobbling something together yourself with all commodity and opensource then you should. However, if you can reach your objectives more quickly or cost effectively by buying a pre-packaged “cloud” offering or using a global public cloud service, then that’s where you should go. The trick is the organizational design. Without proper design you will fail the why process and you’ll be forever fighting ownership disease. And ownership disease can stall or even kill some of your best opportunities because someone’s comfort or skill is being accommodated at a higher priority than the business opportunity.
So get off the cloud bandwagon and get on the actual cloud thought process. Consider what opportunities you’re trying to create and evaluate your cloud or infrastructure solution options based on best fit, as you would with any IT solution. Many of us like to go on for hours about what the pure cloud is supposed to be, but it’s just not an important argument. As long as you have a clear sense of what the technology solution you’re selecting can enable, and it satisfies your requirements, then don’t wait for someone to call it a real cloud, just start using it.
Mark Thiele is executive VP of Data Center Tech at Switch, the operator of the SuperNAP data center in Las Vegas. Thiele blogs at SwitchScribe and at Data Center Pulse, where is also president and founder. He can be found on Twitter at @mthiele10.