VMware has published a paper titled: VMware High Availability (HA): Deployment Best Practices. The paper which contains 10 pages describes the best practices and guidance for deploying VMware High Availability in vSphere 4.1. The paper also includes some discussions about proper network and storage design, and recommendations on settings for host isolation response and admission control.
The paper contains the following sections:
- Introduction, detailing what VMware HA is.
- Design principles for High Availability, by eliminating single points of failure detailing host selection considerations and host placement considerations
- Networking design considerations, by increasing resiliency of “client side” networking to ensure access from external systems to workloads running in vSphere; and increasing resiliency of communications used by VMware HA itself.
- Storage design considerations, by using Multipathing which is a technique that enables you to use more than one physical path that transfers data between the host and an external storage device.
- Understanding host isolation, an ESX host is considered to be isolated when it loses the ability to exchange heartbeat or status information with other hosts in the VMware HA cluster via the service console network.
- Admission Control, which ensures that sufficient resources in the cluster are reserved for virtual machine recovery in the event of host failure.
- Affinity rules, which specifies whether the members of a selected virtual machine Vmware DRS group can run on the members of a specific host VMware DRS group.