HyperV Clustering, Failing Over, and Moving Instances

HyperV Clustering, Failing Over, and Moving Instances

Two common scenarios for higher availability and dynamic datacenters involve:

  • Running the same workload simultaneously on two servers, or
  • Running a workload on a primary server and periodically moving it to a second server due to a failure, load balancing, patching, or planned downtime.

In both scenarios, regardless of whether the workloads are running in physical or virtual OSEs, each server must have the appropriate number of licenses assigned to it prior to the workload running on it. This holds true regardless of whether you plan the workload to:

  • Always run on a single server.
  • Run in parallel on the server as a backup when the primary server fails.
  • Run the workload if the primary server is down.
  • Load balance when the primary server has high use.
  • Only run the workload during maintenance.

Please see the previous section, “License for Peak Capacity.” Figures 2A, 2B, and 2C demonstrate three examples of usage scenarios that are properly licensed.

HyperV Clustering

Example: The servers are clustered, each licensed with Windows Server and both running the same workload in parallel.

HyperV Clustering

Example: The servers are clustered, each licensed with Windows Server and both running the same virtualized workload in parallel.

HyperV Clustering

Example: Both servers are licensed with Windows Server. The workload is moved from the first server to the second server.

© 2014 Pay Per Cloud Professional Services | Sitemap | Legal