Google says that a lightning storm in Belgium last week caused electrical outages to occur at one of its data centers. The Silicon Valley giant is now upgrading some of its equipment after admitting that a small amount of user data was irreversibly lost as a result.
The power grid connected to Google’s facility in Saint Ghislain suffered from “four successive lightning strikes” last Thursday, the tech company said this week, causing problems for machines inside one of its European data centers.
Google said its engineers attempted to recover data through the weekend, but eventually determined that “less than 0.000001%” of the disk space used for certain cloud services had been permanently damaged. The company did not say specifically what data was lost.
“Google takes availability very seriously, and the durability of storage is our highest priority,” the company said on Tuesday. “We apologize to all our customers who were affected by this exceptional incident.”
The Belgium data center, as well as facilities in Iowa and Taiwan, contain equipment that powers the Google Compute Engine, or GCE — a cloud computing infrastructure operated by the Mountain View, California, company that provides services to remote clients the world over.
The power failure only caused permanent destruction to a sliver of the persistent disks in its “europe-west1-b” zone, one of three hosted in St. Ghislain, but it could have potentially caused even bigger problems, Google said.
A spokesman for data center consultants Future-Tech, commented that while data centers were designed to withstand lightning strikes via a network of conductive lightning rods, it was not impossible for strikes to get through.
“Everything in the data center is connected one way or another,” James Wilman, a spokesman for Future-Tech, told BBC. “If you get four large strikes, it wouldn’t surprise me that it has affected the facility.”
Google said in an incident report on Tuesday that it was taking the blame for last week’s outages, but at the same time cautioned users that the cloud is no match for Mother Nature.
“This outage is wholly Google’s responsibility,” the company said. “However, we would like to take this opportunity to highlight an important reminder for our customers: GCE instances and Persistent Disks within a zone exist in a single Google datacenter and are therefore unavoidably vulnerable to datacenter-scale disasters.”
In 2011, a power outage in Dublin prompted data centers operated there by Microsoft and Amazon to go down and cause a loss in service for customers of Amazon’s EC2 cloud-computing platform and Microsoft’s BPOS, Data Center Knowledge reported at the time.
• Andrew Blake can be reached at ablake@washingtontimes.com.
Please read our comment policy before commenting.