Given all the complexities of running a data center, sometimes it’s easy to forget the basics, as the state of New Jersey found out last week. According to ZDNet, the data center outage began in the early-morning hours with a power failure at the utility provider, PSE&G. But following the four-hour power failure, the state’s data center still didn’t return to operation until midafternoon—2 or 3 pm.
The outage highlights two critical preparedness measures that companies (and governments) should implement when running mission-critical data centers: power backup and fault recovery procedures. The first line of defense in a case like this is power backup. In the case of New Jersey’s data center outage, a backup generator, UPS and appropriate power switching could have handled the utility failure—particularly when the duration is a rather meager four hours. Second, for those cases involving unexpected downtime (whatever the reason), recovery procedures developed and practiced ahead of time enable faster returns to uptime. Given that the cost of downtime (maybe more for companies than for governments) can quickly exceed the cost of power backup equipment, organizations can learn an important lesson from New Jersey’s data center outage.
Read more about data center outage