The usual suspects for causing data center outages: failed UPS backups, human error, and cyber crime.
Back in January of this year, the Poneman Institute (@PonemonPrivacy) and Emerson Network Power (@EmersonNetPwr) released their third report on the cost of data center outages. Previously published in 2010 and 2013, the number one take away from their third report is that the cost of down time due to outages is going up—way up. Today’s figure for the average cost associated with a data center outage is $740,357.
You read that right. Nearly three quarter of a million dollars on average, per incident. With the average duration of 95 minutes, downtime in a data center is costing $130 per second. To put it another way, that’s twenty-six Venti Caramel Frappuccio drinks for every second that today’s data center is down.
The new high represents a 38 percent increase from the first time this information was recorded in the 2010 study.
More concerning still is the 81 percent increase in the maximum cost of data center downtime since first publishing this report in 2010. The 2016 maximum cost for data center downtime is a staggering $2,409,991.
The background for this study looked at 69 U.S. data centers ranging in size from 1,505 square feet (139,8 m²) to 55,000 square feet (5 109,7 m²) in 15 different industries ranging in use from eCommerce to healthcare. The costs were based on direct costs, indirect costs, and opportunity costs. The top five contributing cost categories were detection, IT productivity, end-user productivity, lost revenue, and business disruption. The five most effected industry segments were financial services, communications, healthcare, eCommerce, and co-location.
As for unplanned partial outages—while the duration has been relatively consistent across all three study years—the downtime cost has increased significantly. The average cost for a partial outage has increased by 46 percent from 2010 to 2016.
Not surprisingly, the cost per square foot is almost twice as much for the smallest of data centers in the study as compared to the largest data centers.
Reasons for Downtime
The number one reason for unplanned outage remains the failure of UPS systems (25 percent). Tied for second in the 2016 study are human error and cyber-crime (22 percent). Both UPS failure and human error remain consistent with 2013 percentages, indicating that not much has been done to mitigate loss based on seemingly preventable events. Cyber-crime, however, has increase 167 percent from 2010 and represents one of the biggest future threats to modern data center uptime.
The study further breaks down the cost based on the root cause. While UPS failure is the number one reason for downtime, it ranks third in cost. Cyber-crime ranks second in cost, and IT equipment failure is the most costly reason for downtime.
View the full study here: http://www.emersonnetworkpower.com/en-US/Resources/Market/Data-Center/Latest-Thinking/Ponemon/Documents/2016-Cost-of-Data-Center-Outages-FINAL-2.pdf