Specifications

Best Practices for Virtualizing and Managing Exchange 2013

CAU supports an extensible architecture that helps to update the cluster node with node-updating tools

and software updates that are not available from Microsoft or through Windows Update or Microsoft

Update. Examples include custom software installers, updates for non-Microsoft device drivers, and

network adapter/HBA firmware updating tools. This is beneficial for publishers who want to coordinate

the installation of non-Microsoft software updates.

Virtual Machine Priority

IT administrators can configure availability options for virtual machines running on Hyper-V host clusters.

An administrator sets priority for the virtual machines in a host cluster, which the host cluster uses to

identify the high-priority virtual machines and give them first preference. This ensures that high-priority

virtual machines are allocated memory and other resources first upon failover.

In Windows Server 2012, administrators can configure availability options/settings to provide improved

and efficient allocation of cluster resources (such as when starting or maintaining nodes) in large physical

clusters and Hyper-V failover clusters. The availability options for managing clustered virtual machines

and other clustered roles include the following:

 Priority settings: This option can be applied to all clustered roles, including clustered virtual

machines. A virtual machine can be set to high priority, medium priority, low priority, or No Auto

Start. By default, every virtual machine is set to medium priority. Clustered roles with higher

priority are started and placed on nodes before those with lower priority. If a No Auto Start

priority is assigned, the role does not come online automatically after it fails, which keeps

resources available so other roles can start.

 Preemption of virtual machines based on priority: This option can be applied to clustered

virtual machines. In case of a node failure, if the high-priority virtual machines do not have the

necessary memory and other resources to start, the lower priority virtual machines are taken

offline to free up resources. When necessary, preemption starts with the lowest priority virtual

machines and continues to higher priority virtual machines. Virtual machines that are preempted

are later restarted in priority order.

Best Practices and Recommendations

In case of failover, virtualized domain controllers, along with other high-priority workloads like

Exchange 2013 virtual machines, should be given the highest priority to start first.

Virtual Machine Affinity

The Virtual Machine Affinity rules in Windows Server 2012 allow administrators to configure partnered

virtual machines to migrate simultaneously at failover. For example, imagine two machines are partnered:

One virtual machine has front-end applications and the other has a back-end database.

It can also be specified that two particular virtual machines cannot coexist on the same node in a failover

scenario. This is the Virtual Machine Anti-Affinity rule. In this case, Windows Server 2012 Hyper-V migrates

the partnered virtual machines to different nodes to help mitigate a failure. For example, domain