Unplanned data center outages still present a difficult and costly challenge for organizations, according to the results of a survey of U.S.-based data center professionals completed by the Ponemon Institute and sponsored by Emerson Network Power.
Data center outages are so dreaded that 84% of the respondents in the study stated that they would rather walk barefoot over hot coals than have their data center go down. A report detailing the results is available online.
This is the second time Emerson Network Power has teamed with the Ponemon Institute to examine the root causes and frequency of unplanned data center outages. The first study, conducted in 2010, revealed that organizations were underestimating the impact unplanned outages were having on their operations. This year’s report, which analyzes survey responses from 584 U.S.-based professionals involved in data center operations, finds that this is changing, and in some respects, the ability to prevent data center outages is improving.
An overwhelming majority of survey respondents reported having experienced an unplanned data center outage in the past 24 months (91%). This is a slight decrease from the 95% of respondents in the 2010 study who reported unplanned outages. Regarding the frequency of outages, respondents experienced an average of two complete data center outages during the past two years. Partial outages, or those limited to certain racks, occurred six times in the same timeframe. The average number of device-level outages, or those limited to individual servers was the highest at 11. These durations have declined slightly from 2010 findings (complete: 2.5, partial: 7, device level: 10).
Eighty-three percent of respondents said they knew the root cause of the unplanned outage. The top three most frequently cited root causes of outages remain unchanged from the 2010 report: UPS battery failure (55%), accidental EPO/human error (48%) and UPS capacity exceeded (46%). Thirty-four percent of respondents cited cyber attacks, which is up from 15% in 2010, while 30% cited weather-related reasons, which is up from 20% in 2010. Fifty-two percent believe all or most of the unplanned outages could have been prevented.
“As computing demands and complexity in the data center continue to rise, unplanned data center outages remain a significant threat to organizations in terms of business disruption, lost revenue and damaged reputation. However, this report does show that many companies are more aware of the causes of downtime and taking steps to minimize the risk,” said Peter Panfil, vice president, global power, Emerson Network Power. “For instance, the survey results show that those data centers that make it a priority to minimize the risk of downtime and utilize best practices, such as data center infrastructure management (DCIM) and battery monitoring, are able to decrease or virtually eliminate the frequency of those top three root causes.”
Based on the survey responses, below are the attitudes and actions these high-performing data centers have in common.
- Consider data center availability their highest priority above all others, including cost minimization and improving energy efficiency.
- Utilize best practices in data center design and redundancy to maximize availability.
- Dedicate ample resources to bring their data center back up in case of an unplanned outage.
- Have complete support from senior management on efforts to prevent and manage unplanned outages.
- Regularly test generators and switchgear to ensure emergency power in case a utility outage does occur.
- Regularly test or monitor UPS batteries.
- Implement data center infrastructure management (DCIM).
These high-performing data centers experience fewer unplanned outages and a shorter duration of downtime than the overall average. High-performing respondents experienced an average of 1.5 complete data center outages during the past two years compared to an overall average of 2.0. These complete outages lasted an average of 69 minutes, compared to an overall average of 107 minutes. Partial outages occurred 4.6 times and lasted 107 minutes for high-performers, compared to 6.8 times 152 minutes for the average respondent. The number of device-level outages was 5.5 times at 129 minutes for high-performers, compared to 11.3 times at 153 minutes.
“No single technology or best practice can completely remove the risk of downtime,” said Larry Ponemon, Ph.D., chairman and founder, the Ponemon Institute. “However, what this report shows us is that by committing the necessary investment in infrastructure technology and resources and taking a number of actions, organizations can dramatically reduce the frequency and duration of unplanned data center outages that can potentially cost data centers thousands of dollars per minute.”
As with the 2010 survey, this report is the first in a two-part series that examines unplanned data center outages. The second report, timed for release in late 2013, will conclude with detailed information on quantifying the cost of downtime in the data center.
To download the research report and a podcast on the study, or to register for our upcoming webinar “Get the Lowdown on Data Center Downtime,” visit www.EmersonNetworkPower.com/Downtime.