Disaster recovery can be a difficult thing to plan for. You back up your systems; perhaps you replicate your data to an off-site facility; maybe you even build all redundant systems.
After doing all of these things, what is your goal? It is to get your systems back up and running after some sort of a disaster, such as your building burning down; or an electrical failure in your data center.
Below I have listed a few possible scenarios you may encounter. In these scenarios I outline a few problems you may encounter, but from a Server Prospective. The network recovery is for another article.
Lets imagine a disaster recovery scenario. You have four servers, a File/Print Server, Email Server, SQL Server, and a Terminal Server running in your data-center when it burns to the ground. Lucky for you; you had off-site backups of all your servers. You purchase 4 new servers to restore your data to. Unfortunately you were not able to purchase the exact same servers you had before since your old servers were several years old.
In the above scenario; assuming you are running windows 2003 or older, you are going to have a lot of problem restoring your servers. Because of the way Windows 2003, Windows 2000, etc... operates with its Hardware Abstraction Layer (HAL) windows will most likely blue screen after you restore your backups.
Assuming you are running server 2008 or newer, the HAL won’t necessarily be a problem; but you will have to worry about drivers. Tracking down and installing these drivers after restoring your backups is just one more thing you need to worry about in a disaster, which is a time when you are in a hurry to get your systems back online.
In this scenario we have 4 servers, just like in Scenario 1. However, instead of having 4 physical servers, we are running 1 Physical Server, and 4 Virtual Machines. We are backing up our virtual machines using the same method we were using in Scenario 1; and storing them off-site.
Lets assume your data-center burns down like in Scenario 1. In order to get your systems back online you purchase a new server and re-install your Virtualization product, such as VMWare ESX. After installing ESX, you restore your backups. After your backups are restored you start up your virtual machines you power them back up and your systems are up and running again; at least from a server prospective.
In this scenario you have the same configuration as Scenario 2, but you also have an ESX Server located off site. Using a third party product you replicate your data to your office location. In this scenario when your data-center goes offline, recovery is as easy as powering up your virtual machines. at your Disaster Recovery Site.
As you saw in Scenario 1, trying to restore your systems from backup without having identical hardware can introduce some challenges. By using virtualization you eliminate the worries about having to track down the identical hardware; since all your hardware is virtual. As I have shown in the three scenarios above, there are many different ways of achieving the same goal, but virtualization definitely makes your life easier when trying to reach accomplish these goals.