Use Virtual Clustering to Deliver High-Availability Applications
The traditional approach to providing high availability applications to users is to make the application “cluster aware,” that is, capable of calling resources from more than one server, falling back to a secondary server if the production server is offline for some reason. There are two challenges to keep in mind when building a physical cluster that virtualization minimize or eliminates as a concern:
For many organizations, the hardware costs alone make them look for alternatives to running a dedicated cluster in their data center.
Virtualization provides a solution to the hardware cost problem. A virtualized cluster utilizes all the hardware available, unlike a traditional cluster in which 50 percent of the hardware could be idle for most of the time. Moreover, the applications running in a virtual machine do not need to be made cluster aware, because Windows Server 2008 Hyper-V delegates application processes and calls transparently. Any existing application can run as a clustered application in the Hyper-V environment.
Full hardware utilization and no application changes translate into lower capital expenditures on hardware, making more of your resources available to provide better performance to users at all times, not only in a failover situation. The resources on this page have been collected to help explain the concepts and implementation of this type of solution.
Configuring Hyper-V Virtual Machines for High Availability
Hyper-V: Using Live Migration with Cluster Shared Volumes in Windows Server 2008 R2
Planning Hyper-V Virtual Machines for High Availability
Virtual Machine Migration Test Wizard
Microsoft Training and Certifications