Why Your Enterprise Private Cloud is Failing

You've told your ITOps team to make it happen, you've approved the purchase of cloud-in-a-box solutions, but your developers aren't using it. Why?

Forrester analyst Lauren Nelson and myself get this question often in our inquiries with enterprise customers and we've found the answer and published a new report specifically on this topic.
Its core finding: Your approach is wrong. 

You're asking the wrong people to build the solution. You aren't giving them clear enough direction on what they should build. You aren't helping them understand how this new service should operate or how it will affect their career and value to the organization. And more often than not you are building the private cloud without engaging the buyers who will consume this cloud.

And your approach is perfectly logical. For many of us in IT, we see a private cloud as an extension of our investments in virtualization. It's simply virtualization with some standardization, automation, a portal, and an image library isn't it? Yep. And a Porsche is just a Volkswagen with better engine, tires, suspension, and seats. That's the fallacy in this thinking.

To get private cloud right, you have to step away from the guts of the solution and start with the value proposition. From the point of view of the consumers of this service — your internal developers and business users. 

I&O Looks Up at Cloud; Developers Look Down Into It

To them, a private cloud is a service, not an infrastructure stack. They value the speed in which resources can be allocated to them, the simplicity of getting their work done and the lack of friction involved. To get private cloud right you have to start here, and that requires a completely different set of skills — skills your virtualization administrator, frankly, doesn't possess. And honestly, your virtualization administrator probably doesn't see how he benefits from a private cloud. In fact, he's probably threatened by it. 

Cloud developers look at the private cloud through the lens of the public cloud and the benefits it provides: self-service, a low entry price, strictly limited sets of predefined resource configurations, and services and speed — where fast is defined as fifteen minutes or less. Cloud developers want these same characteristics in a private cloud and won't tolerate compromises in autonomy and agility. And the end customer of the private cloud doesn't care at all about the VM container their app will eventually run on or the underlying infrastructure — these elements have all but disappeared for them.

Bottom line: Your private cloud is very different than your static virtualization environment.

In looking at the organizations who are having success with their private clouds, their approach is completely different. And usually starts with a different person in charge — a developer or architect who approaches the cloud, not from the infrastructure up, but as a service. And they start their service definition with the public equivalent. The approach: How can I deliver the same value and experience of the public cloud from within our own data center? Those who have had the most success with this approach have also started with a complete Infrastructure as a Service or Platform as a Service solution, rather than trying to build one up from a virtualization foundation.

So if your team is struggling to deliver a private cloud and you have taken the bottoms-up approach, stop. This is the difficult, culturally challenging and slow approach to cloud value. Instead, put a new cloud admin in charge. Someone who understand and has direct experience with the public clouds, has a service orientation (not an infrastructure orientation) and is willing to start fresh with solutions that meet your developers needs out of the box.   


Wonderfully put. Truth is

Wonderfully put. Truth is that all IT is and always has been a service, but we are only now really starting to learn to think about it that way. The advent of cloud may be forcing the issue. In designing public and private clouds, we definitely need to think from the perspective of designing services.

James - This is an excellent

James -

This is an excellent article highlighting flawed thinking in many private cloud initiatives. However, I don't necessarily agree that I&O-based cloud leaders cannot see this from the right perspective. To do this requires that they listen to their developer/users and take a requirements-led approach instead of an infrastructure-centric approach. I have seen this work in some enterprises.

The bottom line is that no matter who is leading the private cloud initiative, a developer/user-centric perspective and approach is very much a requirement for success. But that's true of any IT project...


Yep! Management is different, too.

John, you're right: savvy I&O leaders CAN do this, but only if they marry the shift in viewpoint James advocates with a shift in how they both build and operate cloud services.

Too often we're seeing I&O teams stuck in the "build" phase, adding incremental VM management tools, catalogs, and portals. They need to be starting with management tools that will give them usage, cost, and service visibility up front - at the cloud service level and not the infrastructure level - if they hope to show users they're running a cloud and not just a quickly-provisioned virtual environment.

Great comment.

If you build it they will come (or not)

I think there is too much focus on packaged solutions to internal private clouds. Even if the figures stack up (including the transformation of current ways of developing), a heap of time needs to allocated to engaging development teams and supporting them through the process of transition.

With the majority of enterprise applications not yet being truly elastic, moving existing applications to internal clouds is not the cost saving bagger that pure visualization is.

If an organization is currently successfully using public cloud services, this is the marker that internal cloud be beneficial. That usage approach should be the blueprint of internal services, and the basis of cost benefit analysis.


Finally someone has spoken out about the right and wrong approach to private clouds. However, this put big question marks on the strategy of major "enterprise" vendors. They didn't get it or they simply pretend not to?

I have written something more about that on my personal blog:

Very true, well put!

Very true, well put!

I totally agree with the

I totally agree with the right and wrong approach on private clouds. Have you looked at the content here provided by HP & Microsoft and their economic analysis: http://hpmicrosoftprivatecloud.com/