One of the primary problems with most approaches to disaster recovery is that it has historically required IT organizations to invest in a whole set of duplicate infrastructure to make sure data is available at any given moment.

But the advent of virtualization is now allowing IT organizations to take a slightly different tack when it comes to disaster recovery by relying more on third-party services. For example, DS3 DataVaulting has launched a Disaster Recovery Solution that leverages a virtualization infrastructure to give customers an approach to disaster recovery that they only pay for based on how much they use.

That means that the DS3 service basically scales on demand, but also allows customers to recover much faster compared to tape backup systems because the DS3 service is leveraging virtual servers and disk-based backup systems.

There's no doubt that IT organizations could replicate the DS3 service themselves. But when it comes to disaster recovery, the one additional benefit that a service provides is that its data center is typically hundreds of miles from the location of the disaster. All too often we see IT organizations creating backup data center facilities that are relatively close to their existing data center operations only to see both facilities overwhelmed by the same event. Furthermore, the biggest issue associated with most disasters is not usually the equipment in the back-up data center, but rather the simple fact that people can't get to the back-up data center for one reason or another.

Over time there will probably be more than a few services leveraging virtualization to provide a different cost model for disaster recovery. So the only real question is when are IT organizations as a whole going to start incorporating these services as part of a core disaster recovery plan that isn't wholly dependent on all the right people being in all the right places at a time when nobody can usually get anywhere.