LAS VEGAS -- As enterprise technologies move from static to dynamic, IT pros must learn new management tactics and rely on new tools.
A successful cloud installment, for example, should not only make the CIO happy, it should also deliver services with an eye on the end user. One way to do that is through automation, said Ronni Colville, an analyst at Gartner, here at Gartner Data Center conference this week.
But many enterprises go about
This increases the management complexity exponentially, Colville said. And while automation tools can help, cloud teams can shoot themselves in the foot by over-specifying cloud automation tools -- fragmenting the management environment.
Choosing automation tools
Deciding which automation tool is best can be a dizzying task, especially in the cloud management platform space -- which comprises about 50 to 70 vendors and is rife with acquisitions and change.
There are benefits to this, though; IT pros have options and the ability to create consistency across the portfolio.
"Cloud forces IT departments to be more standardized," Colville said. Colville encourages IT pros to evaluate tools in a proactive way by going to vendors and selectively pick automation tools that fit their specific requirements. "Pick and choose what end users will get and then pare down tools to what's necessary."
But don't expect to standardize to a single tool. There will never be a single automation solution in your environment, Colville added, so you'll need to know how to manage that.
"If you can get down to four tools, that's good," she said.
According to Gartner, 75% of large enterprises will have more than four diverse automation technologies within their management portfolio by 2015 -- up from less than 20% in 2013.
Opportunistic to systematic: Adjust your automation mindset
Most issues enterprises encounter with cloud automation occur because IT teams take an "opportunistic" approach to management, Colville said, instead of a "systematic" one.
Opportunistic automation, which reckons back to traditional, siloed IT departments, focuses on cost of ownership. IT teams are doing ad hoc automation -- without a central point person saying how much is too much. Just because there's an automation tool for nearly every function, doesn't mean you need it. There's an idea in IT that automation will fix everything. In actuality, too much automation can actually lead to a loss of control, she said.
This siloed approach to automation means networking groups buy the automation tools they need, storage guys buy the tools they need, and so on. Unless the tool is created specifically for collaborative work, everyone has his or her own tool, Colville said. These islands of automation can create inefficiencies and often are not scalable across departments or systems.
Systematic automation, on the other hand, focuses on policies, visibility and governance, Colville said. Automation tools are selected to work across all systems of a data center. To move to this systematic automation mindset, IT teams need to take a step back and look at what they have in an effort to reduce unnecessary or redundant tools.
Michelle Boisvert is executive site editor for SearchCloudComputing, SearchWindowsServer and SearchDataCenter. Contact her at firstname.lastname@example.org.