Is Data Center Consolidation Actually Gaining Steam?

Published: January 30, 2013

Data Center

Could consolidation of federal data centers actually be gaining steam? Extracts from data.gov on the number of planned data center closures indicate that the total planned shutdowns are increasing over time.

Measuring federal data center consolidation progress has become quite complicated.  When Vivek Kundra launched the Federal Data Center Consolidation Initiative (FDCCI), agencies counted nearly 2100 data centers and developed plans to close 800 of them.  When Steve VanRoekel took the helm as federal CIO, he declared a new definition of a data center, bringing the total number of data centers to approximately 2900 with new goals to close 1200. 

 

A recent extract from data.gov indicates that agencies plan to close 315 data centers in FY2013.  That number seemed high to me.  So, I pulled out a previous extract that I pulled back in July and discovered that planned FY2013 closures from that date were only 197.  The chart below shows on an agency by agency basis the difference between the number of FY2013 planned closures from the two data.gov extracts. 

According to the data, the Department of Defense decided to close an additional 44 data centers this fiscal year, Homeland Security an additional 22, and the Department of Justice an additional 13.

 

Why the huge disparity in the number of planned closures?   I can only wager a guess that the difference lies in the definition of a data center.  I would suspect that agencies revised their inventory counts of data centers and subsequently revised their number of planned closures upwards.  Under Kundra’s definition, a data center had to be at least 500 square feet, but under the new definition adopted by VanReokel a data center is

 

“a closet, room, floor or building for the storage, management, and dissemination of data and information. Such a repository houses computer systems and associated components, such as database, application, and storage systems and data stores.' A data center generally includes redundant or backup power supplies, redundant data communications connections, environmental controls (air conditioning, fire suppression, etc.) and special security devices housed in leased (including by cloud providers), owned, collocated, or stand-alone facilities. Under this revised definition, neither square footage nor Uptime Institute tier classifications are required to define a facility

as a data center.”  

 

I suspect as agencies make quarterly updates to data.gov regarding closure plans, they are also discovering and making plans to close smaller data centers not originally identified in the first inventorying effort under Kundra.

 

To validate my suspicions, reviewing updated agency FDCCI plans would be helpful.  However, updated consolidation plans submitted by agencies to OMB at the end of FY2012 have not been released to the public to date.

 

As stated by Steve VanRoekel before the House Committee on Oversight and Government Reform last week, “In essence, we are going to optimize and close data centers by shifting resources of one to other ones – to more efficient data centers- not taking away certain services, not depreciating any service we provide.”  VanRoekel indicated that data center consolidation is more about optimization than it is about data center counts.

 

I agree, but to date it has been difficult to quantify much of the data center consolidation process or its total savings.   This issue was also raised by David Powner of GAO during the same House Oversight hearing.  “Ultimately, it’s about getting to the point where we’re saving money, and that’s where there hasn’t been a lot of transparency at the moment in data center consolidation,” state Powner.  According to Powner, the only agency to proclaim projected savings has been DoD with $2.2 billion in savings due to its consolidation efforts.

 

Expect agencies to continue to consolidate data centers by moving applications and systems to the cloud, implementing shared services, and offloading work and applications to other, more efficient data centers.  Savings will be realized, but don’t expect savings amounts to be clear or the method for counting the number of closures to remain consistent.