One of the best pieces of advice about modern disaster recovery planning can be found in the pages of the 400-year-old literary classic Don Quixote. Paraphrasing an old Spanish proverb, author Miguel de Cervantes notes that a wise man should not “venture all his eggs in one basket.”
It’s a practical message about the risks of concentrating all of your resources in one area, but one that’s too often unheeded by DR planners.
Research suggests that as many as 40 percent of businesses rely upon onsite data center replication for enabling the recovery of vital technologies and data in the event of a disaster. That won’t do any good in the event of a tornado, flood or some other type of natural disaster that incapacitates your main site.
Geographic diversity, or geodiversity, is an essential characteristic of effective DR planning. Replicating data, applications and systems to a site that is geographically separated from your main site helps ensure that a single disastrous event doesn’t take out both locations.
Lessons from Sandy
Even those with a remote site often fail to provide enough distance. Hurricane Sandy in 2012 demonstrated the risk of that strategy. At the time, it was standard procedure for many New York businesses to establish DR sites in New Jersey. However, Sandy devastated areas of both states as well as 22 others, flooding both main and secondary data centers for dozens of companies.
That’s one reason many industry experts suggest that secondary sites should be at least 100 miles away. Companies in areas prone to hurricanes, earthquakes and other types of natural disasters that usually affect large geographic areas should probably push that out to 200 miles or more. However, the facility should also be close enough to deliver applications and data to users with minimal latency.
Distance isn’t the only consideration, however. It may also be a good idea to choose a site that’s on a separate power grid and fiber network from your main location. Even if you’ve safely backed up data to a remote location, you’ll still need electricity, bandwidth and telecommunications capabilities to restore operations quickly.
Of course, setting up a redundant data center in a distant location can be an expensive process. Real estate costs, hardware, software licenses, power, cooling and network connectivity all add up. A standard formula suggests you can count on spending roughly $1,000 per square foot for a data center.
The DRaaS Advantage
There is a far more economical approach that delivers the necessary geodiversity. Cloud-based DR-as-a-Service (DRaaS) solutions provide robust DR capabilities with subscription-based pricing. DRaaS providers typically operate from multiple data center locations to ensure their failover capabilities. For example, RMM Solutions’ DRaaS offering has exceptional failover capabilities through our Type 2 SSAE 16/SOC 1-certified data centers in Wausau, Appleton and Waukesha.
Our solution provides on-demand access to systems and services so there’s no need to purchase hardware or software with DRaaS. In addition to eliminating upfront capital expenses, this model enables a great deal of flexibility by allowing subscribers to spin up new virtualized resources in a matter of minutes. RMM also offers additional services such as DR planning and testing, real-time data replication, data security, and ongoing management and support.
These types of features have driven increased use of the cloud in data protection strategies. According to a recent Unitrends survey, 60 percent of responding organizations report using DRaaS and other cloud services for data backup, storage and archiving. Organizations using DRaaS reported significant reductions in downtime and say they typically can recover failed applications in less than an hour.
“Diligence is the mother of good fortune.” That’s another quote from Don Quixote that applies to disaster planning. Give us a call to learn more about using our DRaaS offering to create the geographic diversity necessary to protect your organization.