(This is the first article in a five-part series.)

If you haven’t gotten around to implementing server virtualization yet, it’s a sure thing you will before long. After all, who can resist a technology that lowers outlays on power and hardware, improves availability, and simplifies disaster recovery? “The case study for virtualization is so compelling that it’s inevitable in the data center,” argues Gregory Ness, vice president of marketing at Cupertino, Calif.-based virtualization security vendor Blue Lane Technologies Inc. “The question isn’t if, but when.”

Still, introducing virtualization is no cakewalk, and newcomers to the technology face significant challenges. Take budgeting, for example. Everyone knows to reserve funds for a hypervisor, the underlying software that makes virtualization work. But a properly equipped virtual environment typically includes these optional elements as well:

New hardware: Most companies buy brawny new servers when introducing virtualization, for two primary reasons. First, host servers with speedy processors and lots of memory usually produce greater energy savings and better performance. Second, some hypervisors work only on servers with CPUs featuring built-in virtualization enhancements, such as the AMD Virtualization (AMD-V) technology in AMD Opteron processors.  Host machines older than 18 to 24 months usually don’t qualify.

Storage: If you don’t have a robust storage area network (SAN) in place already, plan on deploying one in conjunction with your virtualization rollout. Just as virtualization enables multiple servers to share a common host, SANs allow those virtual machines to share common storage resources. Organizations that don’t use both technologies in tandem aren’t fully exploiting virtualization’s power, experts warn.