Expertise is a hot-button issue for the just over 400 respondents to our 2012 Private Cloud Survey, and no wonder. The number of new processes needed to build a private cloud is daunting--think defining runbooks, modifying change control strategies to enable automation, and a fundamental shift in management practices. And you still need new products, configuration management databases, and orchestration platforms to tie everything together.
Recruiting skilled employees was the second biggest challenge for respondents who already have private clouds; those starting to build or shying away from private clouds anticipate problems getting skilled employees.
Automation expertise in particular is a sticking point for a systems manager at a large sporting goods retailer. "This area is still incredibly immature," he says. Things run pretty well up to the point where a server has its OS installed. From there on out, someone has to enable and manage the automated installation of applications, and maintain them once they're in service. "These skills sets are hard to find in a typical server engineering or data center operations staff," he says.
The answer? Look to your desktop and client automation teams, the systems manager says. "They must automate everything they do, since most organizations have thousands more client systems to manage than servers," he says. He expects this skill set to expand into the data center to fill the void.
Are your server, storage, and network admins going to morph into programmers? Probably not. Are your application developers going to sprout administrative chops? No. But everyone will need to work closely at an earlier stage in the application deployment life cycle.
By necessity, a private cloud ends up breaking down silos. That's because to get to an automated state, key staff from each discipline must determine how virtual machines and applications are configured and deployed--before the fact. As for managing the private cloud software itself, the ideal candidate will be familiar with the programming languages being used to integrate various components. For example, Microsoft's Azure relies heavily on PowerShell. Python and Ruby are also popular.
Breaking down silos is just one area where the 21% of respondents with clouds in place have had to change the way they do business. It's also a key lesson for the 30% starting down the road.
Payoffs In Sight
So is a private cloud move worth it? To find out, we took a new approach with our Private Cloud Survey, breaking it into three parts. We sorted respondents based on whether their companies had a private cloud in place, were in process building one, or didn't have plans for one. Then we asked specific questions of each group. In our full report, we highlight differences between expectations and reality, and highlight lessons learned from respondents already using private clouds.
What we found was that those with private clouds use hardware more efficiently, report superior scalability and reliability, and make better use of IT's time. Adopters also report lower capital and operational costs. So it's ironic that budget is the biggest reason cited by respondents who aren't planning to implement private clouds. They say either the IT dollars are allocated elsewhere or there's no money available for these projects.
How best to get these hesitant companies to free up some resources: Show proven operational cost savings and clear models that IT can relate to that demonstrate the business advantages. Sorry, pointing to Google's claimed 1:1,000 admin-to-server ratio isn't going to cut it.
Google in the Enterprise SurveyThere's no doubt Google has made headway into businesses: Just 28 percent discourage or ban use of its productivity products, and 69 percent cite Google Apps' good or excellent mobility. But progress could still stall: 59 percent of nonusers distrust the security of Google's cloud. Its data privacy is an open question, and 37 percent worry about integration.
InformationWeek Tech Digest, Nov. 10, 2014Just 30% of respondents to our new survey say their companies are very or extremely effective at identifying critical data and analyzing it to make decisions, down from 42% in 2013. What gives?