Those enterprises with cloud computing strategies have already started the journey toward the cloud. This typically means building cloud-based systems, usually infrastructure, that initially supports a small amount of use. However, the idea is to build these clouds into bigger and better things.
Most often these “clouds with training wheels” are private clouds, which still seem to be the clouds of preference within larger enterprises and the government. This is mostly around comfort levels than capabilities, more about control than compliance.
The trouble is, many enterprises that are working on their first cloud instances are actually dealing with “cloud platforms” that are not “cloud platforms” at all. These are typically virtualized systems that both the technology provider and those who consume this technology call “a cloud.” This type of virtualized technology is as much a cloud as my first bicycle was a Formula One racer.
The problems arise when you try to pin down the definition of a “cloud.” There is no clear, solid, and defendable definition. This is an old story. Every vendor has positioned their technology in the cloud space, no matter if it resembles a cloud or not.
So, those in enterprise IT may believe they are buying a cloud when it’s really a fake. A fake cloud cannot provide them with all of the features and advantages of true cloud-based platforms, which include, but are not limited to:
- The ability to instantly self-provision resources.
- The ability to instantly auto-provision resources.
- The ability to scale to the functional limits of the enterprise.
- The ability to align costs with use.
- The ability to manage the interactions with other systems, including leveraging heterogeneous cloud-based systems.
Fake clouds don’t support these features, and are typically simple virtualized systems that are a bit more efficient than stand-alone servers, but they don’t get anywhere near the objectives and value of cloud computing. However, those who have deployed those systems have been calling their systems “cloud” for so long that it’s become the lie that gets repeated so many times that it seems like the truth.
While some would call private clouds fake clouds, the reality is, that’s not true either. If clouds are able to provide the features and advantages defined above, then many private clouds are certainly clouds. They are just clouds that exist in your own data center, or sometimes under people’s desks. However, they scale, they provide provisioning, they align use to costs, and thus they are a true cloud.
The problem with private clouds is that they look exactly the same as fake clouds, and thus why they are difficult to distinguish. Private clouds are also collection of servers that are working together leveraging the features and advantages listed above. Fake clouds are also a collection of servers, but without the advantage of auto- and self-provisioning.
The use of public cloud computing, while more obvious, can provide even more confusion. Many enterprises call anything running outside of their firewall a “cloud.” In reality, unless it supports the features and advantages listed above, it’s typically fake as well.
At issue is that most enterprises are cultivating fake clouds as real clouds. They have no idea that what they are working with won’t have the value that true clouds are able to provide. They won’t understand that fact until things in the emerging cloud computing industry become more obvious. But, at some point, it will be understood and you need to be prepared to defend your use of fake clouds, or do a quick and silent switcheroo.
If you’re in a Global 2000 enterprise, chances are your initial movement into the world of cloud computing involved the use of fake clouds, at least at first. Typically, you do not understand that you’re cultivating a fake cloud, and perhaps it’s time you found out.
So, what does an enterprise using fake clouds do to go legit?
First, it’s time to define the true value of cloud computing for your specific enterprises, including what value cloud computing will provide to the business. You quickly get to agility and time-to-market benefits, which are driven by auto- and self-provisioning, as well as elasticity, things that are not available in fake clouds.
Second, set a path to true cloud computing using the right technology for the job. This means moving away from simple virtualization as a substitute for a cloud. Virtualization does have its place, but it’s not a cloud. Thus, this also means understanding the differences between fake and real clouds. Finally, being able to test and deploy the real cloud platforms appropriately…as clouds.
Finally, it’s a good idea to reevaluate the culture within IT. Fundamentally, the idea of cloud computing is about thinking differently in terms of how IT consumes IT resources. Those who don’t get that fact need to be retrained. Those who can’t be retrained should be replaced.
So, all is not lost if you’ve discovered you’re cultivating fake clouds. However, the first step is to understand you have problem. The rest is pretty easy.