3 Lessons Learned from Feds' Data Center Consolidation

The ongoing effort to reduce the federal government’s inventory of data centers appears stymied by changing definitions, a lack of resources, and skepticism. The troubled project provides a cautionary tale for IT executives facing their own data center consolidation initiatives.

Testifying before a House subcommittee late last month, federal CIO Steven VanRoekel told legislators that his expansion of the definition of what constitutes a data center is the reason that the federal data center count now stands at more than 7,000.

That’s 5,000 more than were thought to exist when former federal CIO Vivek Kundra first proposed the so-called Federal Data Center Consolidation Initiative (FDCCI) in 2010.

Not that there hasn’t been progress: VanRoekel says 484 federal data centers have been shuttered since the effort began, and he claims another 855 are on track to close by September 30, the end of the federal fiscal year.

But the fact that the data center count keeps growing has some observers wondering whether the program can succeed, or whether it’s even the right effort for a federal government that’s saddled with swaths of aging legacy systems and excessive duplication of applications -- challenges that aren’t exactly foreign to large corporations.

“It’s not just about just data centers,” said Steve O’Keeffe, founder of MeriTalk, a public-private partnership that works to promote more efficient government IT. “The data center element is just a symptom of the challenge.”

O’Keeffe knew the challenges facing FDCCI were worsening when he testified before the House subcommittee in May and no one from either the Office of Management and Budget (which includes VanRoekel) or the General Services Administration showed up.

“To have a hearing on federal data centers and cloud computing without anyone from OMB or GSA is like Disneyland without Mickey Mouse,” O’Keeffe said.

Still, VanRoekel’s presence at that initial hearing would not have changed the big picture. The way O’Keeffe sees it, FDCCI is destined to fail for a number of reasons, ranging from the aforementioned growing data center count and a lack of resources at the agency level to shrinking morale and an ill-advised strategy.

[Taking a software and data-driven approach can help reduce costs and maintain availability as IT becomes more complex. Read about one of the companies in this emerging area in "Tackling Information Infrastructure Complexity."]

Many of the issues FDCCI faces should ring familiar for enterprise IT leaders, who often are saddled with underfunded, poorly thought out mandates that chip away at morale.

In fact, a couple of the ways O’Keeffe said FDCCI has gone wrong would seem to offer up lessons to such besieged IT executives:

•While one of the main drivers for the federal effort was a desire to reduce electricity bills, which account for 12 percent of data center costs, no one has been able to pinpoint how those costs break down for each data center. “If you don’t know how much it costs to power each data center, how do you know which ones to consolidate?” O'Keeffe asked.

Lesson for IT leaders: Figure out which data center resources are power hogs so you know what needs to go.

•Application sprawl must be addressed. According to the Government Accountability Office, federal agencies are running 777 separate supply chain systems and 620 different human resources systems. O’Keeffe said the U.S. Army runs more than 100 operating systems in its data center to support an array of legacy applications. “If you can’t close down an application, then you can’t close down a data center," he said.

Lesson for IT leaders: Start by aggressively rationalizing your application portfolio lest your consolidation effort leaves you with an over-taxed infrastructure.

To understand just how wrong a data center consolidation can go, consider the findings of MeriTalk’s recent survey of 200 federal data center operators. Not only did 56 percent give their agencies’ consolidation efforts a grade of C or below -- a clear sign of a project headed south -- a whopping 45 percent believe that it’s more likely they’ll win the lottery, the Washington Redskins will win the next Super Bowl, or another large meteor will hit the Earth than agencies will meet FDCCI objectives.

Far from a glowing endorsement, but it serves as a third lesson -- an important reminder to IT execs that they’d be well advised to consider their data center teams’ perspectives before going full steam with consolidation mandates.