Understanding Hyper-V and Virtual Machines in the Context of a Cluster

  • [アーティクル]

Applies To: Windows Server 2008 R2

This topic provides information about the following:

  • Overview of Hyper-V in the context of a failover cluster

  • Using Cluster Shared Volumes with Hyper-V

  • Live migration, quick migration, and moving of virtual machines

  • Coordinating the use of Hyper-V Manager with the use of Failover Cluster Manager

Overview of Hyper-V in the context of a failover cluster

The Hyper-V role in Server Manager enables you to create a virtualized server computing environment in which you can create and manage virtual machines that run operating systems, applications, and services. Failover clusters are used to increase the availability of such applications and services. Hyper-V and failover clustering can be used together to make a virtual machine that is highly available, thereby minimizing disruptions and interruptions to clients.

You can cluster a virtual machine and you can cluster a service or application that happens to be running in a virtual machine. If you cluster a virtual machine, the cluster monitors the health of the virtual machine itself (and will respond to failures by restarting the virtual machine or failing it over). If you cluster a service or application that happens to be running in a virtual machine, the cluster monitors the health of the service or application (and will respond to failures by restarting the application or failing it over).

A feature of failover clusters called Cluster Shared Volumes is specifically designed to enhance the availability and manageability of virtual machines.

Using Cluster Shared Volumes with Hyper-V

Cluster Shared Volumes are volumes in a failover cluster that multiple nodes can read from and write to at the same time. The Cluster Shared Volumes feature is only supported for use with Hyper-V (a server role in Windows Server 2008 R2) and other technologies specified by Microsoft. For more information about Cluster Shared Volumes, see the following topics:

Live migration, quick migration, and moving of virtual machines

With failover clusters, a virtual machine can be moved from one cluster node to another in several different ways: live migration, quick migration, and moving. This section describes these actions. For information about how to perform the actions, see Live Migrate, Quick Migrate, or Move a Virtual Machine from Node to Node.

Note

If you want to cluster virtual machines and use live migration or quick migration, we recommend making the hardware and system settings on the nodes as similar as possible to minimize potential problems.

The following list describes the choices:

Live migration: When you initiate live migration, the cluster copies the memory being used by the virtual machine from the current node to another node, so that when the transition to the other node actually takes place, the memory and state information is already in place for the virtual machine. The transition is usually fast enough that a client using the virtual machine does not lose the network connection. If you are using Cluster Shared Volumes, live migration is almost instantaneous, because no transfer of disk ownership is needed. A live migration can be used for planned maintenance but not for an unplanned failover.

Note

You cannot use live migration to move multiple virtual machines simultaneously. On a given server running Hyper-V, only one live migration (to or from the server) can be in progress at a given time.

Quick migration: When you initiate quick migration, the cluster copies the memory being used by the virtual machine to a disk in storage, so that when the transition to another node actually takes place, the memory and state information needed by the virtual machine can quickly be read from the disk by the node that is taking over ownership. A quick migration can be used for planned maintenance but not for an unplanned failover.

You can use quick migration to move multiple virtual machines simultaneously.

Moving: When you initiate a move, the cluster prepares to take the virtual machine offline by performing an action that you have specified in the cluster configuration for the virtual machine resource: Save, Shut down, Shut down (forced), or Turn off. Save (the default) saves the state of the virtual machine, so that the state can be restored when bringing the virtual machine back online. Shut down performs an orderly shutdown of the operating system (waiting for all processes to close) on the virtual machine before taking the virtual machine offline. Shut down (forced) shuts down the operating system on the virtual machine without waiting for slower processes to finish, and then takes the virtual machine offline. Turn off is like turning off the power to the virtual machine, which means that data loss may occur.

The setting you specify for the offline action does not affect live migration, quick migration, or unplanned failover. It affects only moving (or taking the resource offline through the action of Windows PowerShell or an application). To specify this setting, see the "Additional considerations" section in Live Migrate, Quick Migrate, or Move a Virtual Machine from Node to Node.

Coordinating the use of Hyper-V Manager with the use of Failover Cluster Manager

After you configure clustered virtual machines, you can modify most settings of those clustered virtual machines using either Hyper-V Manager or Failover Cluster Manager. We recommend that you use Failover Cluster Manager for modifying those settings, as described in Modify the Virtual Machine Settings for a Clustered Virtual Machine. If you decide to use Hyper-V Manager to modify virtual machine settings, be sure to open Failover Cluster Manager and refresh the virtual machine configuration, as described in Refresh the Configuration of a Virtual Machine.

Additional references