Just Google it! “Data center downtime” and the words of your choice: “crazy,” “bizarre,” “annoying,” or “dumber than dumb.” All sorts of off-the-wall events can hit you with the sudden loss of critical systems. And it’ll probably happen at the worst possible moment.
It is time to enable near-zero downtime if you haven’t already done so. There are obvious reasons why. This bit of Googling will remind you that there are less than obvious reasons, too.
Downtime happens when:
Squirrels have it in for you.
No kidding. There were 137 power outages caused by squirrels in 2015. A single squirrel took out half of a Yahoo data center in 2010 when it gnawed through a communications line.
You mess with atomic clocks.
Adding a leap second to Universal Time shut down LinkedIn, Mozilla and Reddit in 2012.
Hunters use you for target practice.
Hunters have a habit of using fiber optic cable for gun target practice. At least they did when it was hanging from power poles near Google’s data center in The Dalles, Oregon. The cables are now safely underground.
An anchor snags an undersea cable.
The world as we know it depends on undersea mega-cables that carry Internet traffic from continent to continent. In 2008 an anchor snagged one and sliced service to parts of the Middle East and Asia.
A curious soul presses a big red button.
A Charlotte data center went down in 2000, costing millions in revenue. Why? A bored “junior consultant” discovered a big red button — the emergency power shutoff — and just had to push it.
When a server fails, you need it restored in minutes rather than hours, in order to minimize the impact to the business. Otherwise you can score one for the squirrels.
The good news: With the right backup and recovery technology in place, data and applications on non-system volumes are instantly available and accessible during recovery.
Our Live Recovery technical brief that shows how to recover from a server failure in minutes — in a few simple steps.