Best of Industry Perspectives: Why Data Centers Are Slow to Adopt Microgrids (And Why That Will Change)
In this edition of Industry Perspectives, we present one of the most popular columns on Microgrid Knowledge during the first quarter of 2018. AFCOM’s John Heiderscheidt explains why an increasing focus on resiliency is pushing more data centers toward microgrids.
Whether you’re surfing the web at work, shopping for a new dress on your tablet or paying those pesky utility bills with your app, you’re utilizing a company’s IT system. That company has chosen to house their network and data somewhere – an on-prem data center, a colocation facility, or increasingly, a cloud vendor like AWS or Azure. Without that company’s servers humming along, you’d have to mail your check to the utility company, head to Nieman Marcus for your garb, or (gasp!) miss out on The Verge or WIRED during your 15 minutes of lunch time.
Retail e-commerce sales in the U.S. accounted for $322.17 billion dollars of the GDP in 2016 according to Statista. That number grew to $453.6 billion dollars in 2017 according to the U.S. Commerce Department. This hardly takes in to account all of the useful (and sometimes useless) traffic going across the country’s intricate network of fiber lines, computer servers, and energy grids. All of it flows in and out of data centers. And when data centers go down, companies lose money.
Data center operators talk about things like redundancy and resiliency almost ad nausea, but for good reason. A redundant data center offers multiple methods of keeping the power on in the event of an outage in the infrastructure ecosystem. A resilient data center rebounds quickly after that dreaded outage hits. Microgrids offer the ultimate in resiliency and redundancy — power available to your data center even in the event the electrical grid goes down. Of course, this luxury comes at a cost.
The capital expenditures required to build a microgrid, particularly for the data center, can be high. Microgrid development costs fluctuate widely, but for a data center it’s easy to cross the million dollar threshold. Data center facility managers love resiliency and redundancy; their financial backers usually hate the phrase “CAP EX” just as much. As with most things in life, it’s a balancing act. Today a data center microgrid may be viewed as a luxury, but does that characterization really hold water?
In a world more threatened than ever by intense natural disasters, or the risks of man-made disasters, the microgrid is a cost of doing business for some organizations.
Consider the British Airways disaster of mid-2017. The airline’s data center went down for three days between May 27, 2017 and May 30, 2017. The outage cost $102 million, and stranded 75,000 passengers, many of whom would likely look elsewhere for their next flight booking experience. Of course not every hit is that bad. Emerson Network funded a study suggesting the average cost of downtime being a more modest $730,000 per outage.
People and businesses rely on connectivity the same way they rely on electricity, water, and even oxygen. Instantaneous modes of communication and commerce are far more than amenities of a modern age. They are major components of our national and global commercial ecosystem. In a world more threatened than ever by intense natural disasters, or the risks of man-made disasters, the microgrid is a cost of doing business for some organizations. Hospitals and Universities have been some of the early adopters of microgrid solutions, for much of the same reason data centers will begin to adopt them — in a connected world, there’s just no room for downtime.
John Heiderscheidt is the president of the Chicago AFCOM Chapter, and the chief compliance officer for MDI Access.
