A virtual natural disaster slammed campus over spring break as the Division of Information Technology simulated the loss of the University of Maryland’s primary data center.

The drill on March 17 and 18 had nearly three dozen staffers scrambling to ensure that the campus’s IT infrastructure could keep running in the event of a fire, hurricane, flood or whatever else Mother Nature could throw at us.

“The team has three to four hours from the time we start this to demonstrate that all the critical applications are working in our backup data centers,” said Jeffrey K. Hollingsworth, UMD vice president and chief information officer.

The applications they restarted from backups include ones vital to the university’s financial and payroll activities, student records and human resources databases.

The exercise generally went according to plan, with a few “interesting” diversions from the expected, Hollingsworth said.

“I don’t score this on how many things go perfectly the first try,” he said. “I look at when something doesn’t go perfectly on the first try, how well and effectively can you get around it.”

Hollingsworth offered a three-tiered plan for making sure you don’t lose data in case of a fire, flood, burglary or other series of unfortunate events—even if you’re cut off from Wi-Fi or the cloud:

  • Have a physical backup, and keep it in a different building. “This goes for everyone with a laptop the same as it does for a manager of an IT system.”
     
  • Maintain a healthy skepticism. “If the software says it’s backing up, trust but verify. Try to restore a few files to make sure it’s working.”
     
  • Whether in an large organization or two-person household, “Make sure more than one person knows how to do the backup and restore. You need redundancy in your people, not just your technology.”