NEW YORK-The days of virtualization being used only as a server consolidation tool are now past.
At the IDC Virtualization Forum here, the research firm’s analysts portrayed x86 virtualization technology as beginning to move well beyond basic server consolidation and software development to spread into new areas that include disaster recovery, high availability and business continuity.
During his May 7 keynote address, John Humphreys, an analyst with IDC and a close observer of this field, said he eventually sees x86 virtualization as a key to such developing enterprise technology as SOA (service-oriented architecture) and cloud computing, where resources are pooled and applications are treated as appliances.
While such lofty predictions for virtualization are common, the field is still very much in its infancy, although there are signs of its rapid expansion and importance within IT. IDC said it believes that 22 percent of installed servers are virtualized and that in the next 12 to 18 months that number will grow to 45 percent.
In addition, more and more ISVs are developing software that works with and takes advantage of virtual environments. Humphreys cited SAP’s certification of VMware in December 2007 as an important step in showing that mission-critical applications can work within a virtual machine.
In the coming year, Humphreys said he believes that the most important part of virtualization will be what he called mobility or “live migration.”
This aspect of virtualization allows an IT administrator to move a virtual machine from one physical server to another or to move a number of virtual machines around an environment. VMware, which is still considered the leader of x86 virtualization, allows this through its VMotion feature, while Citrix and Microsoft have added or will add similar features to their virtualization management suites.
For enterprises, this type of live migration feature not only allows for disaster recovery and high availability, but allows IT to manage planned downtime, which translates into better business continuity and better way to perform maintenance functions such as patching.
The advantages of using virtualization for business continuity and disaster recovery are numerous, but the main reason is the complex and interconnected nature of current systems and the lost revenue and assets that could result from a failure in one part of that network of systems.
The move into providing disaster recovery and business continuity also means that virtualization vendors and their channel partners have a way to enter the lucrative small and midsize business market, where IT budgets are small but the potential loss of a day’s work could wreck a business.
The next step is to ensure that the disaster recovery abilities of virtualization at the hypervisor level move up and down the entire stack. While the hypervisor has good control of the hardware and can help in the case of a physical system failure, the technology has limited capacity to address failures at the operating system and application level.
While companies such as IBM and Oracle can monitor the application environment, it’s virtualization that can allow administrators to change the IT infrastructure to meet the challenges of the day.
“Where I see an opportunity for the industry and how it benefits us as customers is if I can tell you what is going on at the application level and measure that against a set of policies that my organization has set up, I can make recommendations on how to change the infrastructure,” Humphreys said.
“This is where virtualization comes in,” he continued. “Because you can resize a virtual machine or dynamically provision a new virtual machine and then move it to a host that has more capacity, you have the ability to change.”
Eventually, this type of infrastructure will allow businesses to move applications into the cloud, although IDC analysts said they believe this type of change is at least four years away from being a reality.
While Humphreys’ talk focused mostly on the upbeat nature of the field of virtualization, he noted that there remain numerous obstacles to overcome, including virtual server sprawl, how to decommission virtual machines when they are no longer needed and how to maintain security as IT departments move these virtual machines along a company network.
