To make the most of “private cloud” depends on clear understanding of the technology, and how it applies to your own organization.

That kind of teamwork was the advice of “Customize your APM” earlier this week, and it applies with at least equal force to the subject of private clouds. That’s the message, at least, of two recent reports from analysts.

In “Cloud Adoption & Buyer Information“, Michael Steinhart, Editor in Chief of theitpro, reports on a poll he conducted at a recent Midwest USA information technology (IT) conference. Most respondents were either “… building a [public-private cloud] hybrid environment” or “… virtualizing some of our infrastructure”, with another meaningful slice “… experimenting with public cloud services.” For this audience, at least, a full-blown focus on private-cloud buildout is a minority concern. That doesn’t mean a private cloud is wrong for you; it does mean most of the marketplace has its attention elsewhere. If you’re the belt-and-suspenders sort of leader who prizes the flexibility of public-private combinations, or your organization still is learning the benefits of virtualization, you’ll find plenty of help and opportunities from colleagues and vendors. If you have already decided on a private cloud for its compliance, cost, or other advantages, recognize how much research you’ll do on your own.

A good place to start any research, incidentally, is with “Gartner Highlights Five Things that Private Cloud Is Not“. I’ve often been critical of the methodologies of Gartner analysts and related “white papers”; “… Five Things …”, though, is a modest and largely accurate contribution. Its real value is not to help you pass a certification test or pick at the nits of your speech. Instead, examination of misconceptions and confusions significantly helps you clarify your own goals and assumptions.

Consider, as an example, the Gartner author’s third corrected misconception: “private cloud is not necessarily on-premises”. Just in design of the privacy of a private cloud, an information architect has several different dimensions to control, including physical location, ownership, management responsibility, and security of different sorts. A US organization might choose to host its private cloud, for instance:

• on its own premises, to leverage the sunk costs of its existing equipment, and have assurance that certain data have never intentionally been transmitted outside the firewall;
• in a nearby shared-tenancy datacenter, to gain the advantages of professionalized physical security and redundancy;
• in a leased facility in Canada, linked by virtual private network, for legal and tax benefits; or
• any of a number of other combinations of technologies, ownership, and financing.

No one choice is right for all organizations. Your opportunity and responsibility is to analyze your own precise requirements, and judge what combination of technologies and business arrangements
fulfills them.

Sweeping away misconceptions is timely in a cloud market already afflicted with sloppy language. When you re-fuel at a self-serve gasoline station, you probably have only two or three selec‍tions from which to choose. IT is not like that. Cloud purchase is far more complex than filling your fuel tank, with the potential for a correspondingly large reward. It’s essential to have an accurate picture of your target, one you can express in words that make sense both to colleagues and providers.