According to CloudTweaks, RightScale reported that there were 27 notable publicly-reported cloud outages around the world last year, with six of them actually caused by “super storm” Sandy.  “Based on the RightScale information, 26 percent of outages in 2012 were private data centers, another 26 percent were public clouds, 7 percent were SaaS offerings, and 41 percent were hosting providers.”

“The biggest causes included power loss or failed backup (33 percent), natural disaster (21 percent), traffic and DSN routing (21 percent), software bugs (12 percent), human error (6 percent), failed storage systems (3 percent) and network connectivity issues (3 percent).”

The average time to recover was about 7.5 hours, with the obvious impacts on businesses that leverage these services.  This ranged from minor inconveniences, to businesses not being able to operate at all, and with the assistance offered by cloud computing providers ranging from exceptional to non-existent.

Power problems will always be a way of life for most cloud providers, even with power resiliency programs and technology in place.  The challenge will be to reduce the impact of wide spread power disruptions on cloud providers, including more proactive monitoring of the weather and providing higher priced services for those that need more resiliency.