All of this is followed by the latest report on the effects of climate change in 2011. Per a July 10 Reuters report (by Deborah Zabarenko), “Climate change increased the odds for the kind of extreme weather that prevailed in 2011, a year that saw severe drought in Texas, unusual heat in England and was one of the 15 warmest years on record, scientists reported on Tuesday. Overall, 2011 was a year of extreme events—from historic droughts in East Africa, northern Mexico, and the southern United States to an above-average cyclone season in the North Atlantic, and the end of Australia’s wettest two-year period ever—scientists from the U.S. National Oceanic and Atmospheric Administration and the United Kingdom’s Met Office said.”
This is not the apocalypse, it is the new normal. Have you planned for it?
In planning, designing, and operating mission-critical facilities there is no end to the list of disasters we have to prepare for. However, nature’s wrath is one that even those most dedicated to uptime have found challenging. That lesson was learned on July 2, 2012, during the latest outage at Amazon’s data center in Ashburn, VA. As reported by Wired Cloudline and The Seattle Times, “a storm took out Amazon’s backup generators” causing service interruptions for Netflix, Pinterest, Herku, and Instagram.
If storms can take out an uptime-dedicated organization like Amazon, they can certainly bring down less protected data centers.
Point in fact, according to the “2010 National Survey on Data Center Outages” written by the Pomemon Institute, an independent research firm located in Michigan (report sponsored by Emerson Network Power), most of you agree. According to the survey (453 respondents):
• Sixty-eight percent feel they do not use best practices in data center design and redundancy to maximize availability
• Fifty-nine percent do not believe that their power is good and free from sags and surges
• Fifty-eight percent said that data center availability is not their highest priority
Perhaps the most striking part of the survey is that of the top seven causes of data center unplanned outages weather did not even make the list:
• UPS battery failure
• UPS capacity exceeded
• Accidental EPO/human error
• UPS equipment failure
• Water incursion
• Heat-related/CRAC failure
• PDU/circuit breaker failure
Are we so focused on the failures of the infrastructure we put in place to prevent data center outages that we are missing the bigger picture of how do we protect ourselves from the ravages of nature?
Is anyone prepared for:
• A wild fire to engulf their facility?
• Five, 10, 20 inches of rain to fall in one to 24 hours?
• An extended drought limiting water availability for cooling systems?
• Their data center and the surrounding geography to be without utility power for a week or longer (168 hrs+) ?
While these events may not be as frequent as the failure of the equipment components protecting our data centers, when weather strikes it usually is devastating and enduring. So what will you do when Mother Nature comes knocking on your data center door? Will you have a plan?