Just about everybody has some sort of disaster recovery plan in place, and most IT organizations have embraced virtualization to one degree or another. But one thing most IT sites have yet to do is tightly couple virtualization to their disaster recovery strategy.
Hurricane Bill didn't make landfall in the U.S., but it should serve as a wake-up call to IT organizations that have disaster recovery strategies based on what are now archaic technologies. The advent of virtualization technologies means that not only is the recovery time decreasing in the event of a disaster, but the costs are dropping as well.
Instead of having to set up complete duplicates of various mission-critical servers, IT organizations can now quickly replicate a given set of applications and related software infrastructure in a matter of hours. This significantly reduces the hardware costs that have been one of the biggest limitations that IT organizations have faced when setting up a disaster recovery plan. After all, getting business executives to pay for additional hardware that stands idle unless there is an emergency is a difficult concept to sell, even in the best of times.
But according to a study conducted by the National Archives and Records Administration, 93 percent of the businesses that lost a data center for 10 days or more wound up filing for bankruptcy. The challenge has been how to create a truly effective disaster recovery strategy without breaking the bank.
Virtualization software from companies such as VMware, Citrix, Microsoft and others changes the fundamental economic of disaster recovery because not only is it easier to provision a backup server, IT organizations can also now dynamically move application workloads automatically. That means that should a hurricane approach a data center, the IT organization can quickly move the affected application workloads to another data center hundreds of miles away. And with the advent of various cloud computing services based on virtualization, those backup servers can either reside inside the IT organization or be a set of cloud computing services made available by any one of a thousand hosted service providers.
When it comes to virtualization, most IT organizations are still a little too obsessed with the benefits of server consolidation and increased utilization rates. What many of them have yet to do is explore the inherent distributed computing attributes of virtualization.
Unfortunately, even after events such as Hurricane Katrina, most IT organizations only visit their disaster recovery strategies once a year at best. And Murphy's Law will dictate that those reviews will never happen just in time to meet the next major crisis. So as we enter the hurricane season for 2009, if you have yet to revisit your disaster recovery strategy in light on new advances in virtualization technologies, don't be surprised when investors in the business starts questioning the competency of the IT department if everything isn't back online in 12 hours or less.
So the real question is, do you really know if your organization have a thoroughly tested state-of-the art disaster recovery strategy based on virtualization, or are you just hoping it does?