How Data Center Site Selection Impacts Disaster Recovery Efforts

The most current disaster on the public’s mind, of course, is COVID-19, the global pandemic. That said, the rate of natural disasters, such as hurricanes, earthquakes, tornados, floods or wildfires, has also increased in recent years. A new report from Databank highlights the importance of disaster recovery capabilities in such an environment.

Disaster Recovery as a Service (DRaaS), according to the report, is a way for enterprises to manage the risks associated with such disruptions.

“With a DRaaS solution, your servers and data are continuously replicated in an alternate location or data center enabling your IT operation to ‘failover’ in the event of a catastrophe that forces evacuation from your primary site,” Databank reports.

That said, a failover location needs to be able to provide what the company calls a “degree of geographic diversity.”

There are many factors to consider when evaluating geography and placement. These factors work to determine how “well-protected” your data will be, Databank points out, and related, how easy it is to recover from a disaster.

The first factor to consider when designing or selecting a disaster recovery solution? Physical distance.

“How far away should the failover facility be from your production site?” asks Databank.

According to the report,  The general rule of thumb is that the recovery site should be “far enough away to be safe,” so that it  will not be subject to the majority of the same risks as your primary site.

The decision will likely come down to whether to pick a location within the same metro area as your primary site, or a different region all together. This will largely be impacted by the disaster risk in you current location, and/or the nature of your application or data needs.

The report offers this example: For a low-priority application that needs to be protected against a relatively localized threat, an alternate facility in the same metro area (within 30 miles) may be sufficient.

“For instance, a SaaS provider in Manhattan, concerned about maintaining a DevOps platform in a terrorist attack, could effectively failover to a facility across the Hudson River in New Jersey, just a few miles away and unlikely to be affected by the same event,” the report states. ” But a healthcare information provider in Miami, with a mission-critical patient application that needs to be operational in the event of a hurricane, may need a facility in another region entirely (250 plus miles). In that scenario, Atlanta may be ideal as it sits farther in-land than Miami and would be unlikely to experience catastrophic effects of the same storm.”

Get the full report, “Far Enough to be Safe, Close Enough to be Fast: A Field Guide for Adding Geographic Diversity to Your Disaster Recovery Plan,” to learn more about how data center site selection can make or break your disaster recovery plan, as well as Databank’s own geographic diversity.