What is the best way to introduce a second hypervisor? Should it be used in production?
There is no technical reason that any mainstream hypervisor should not be suitable for a production data center, but organizations that mix hypervisors in production are rare. This is mainly due to limited IT staff familiarity with multiple virtualization platforms -- it's costly and time-consuming to train everyone on multiple hypervisors. For example, if there's an issue or problem with a server using ESXi but the IT administrator on duty only knows Hyper-V (or vice versa), the IT staff cannot respond adequately to production needs.
When a second hypervisor is deployed in a production environment, it is typically used in limited deployments for secondary or non-essential workloads. So, if there is a delay in provisioning or resolving problems, the impact on production is small or non-existent. In many cases, the second hypervisor is used outside of production in a test and development environment.
Alternately, a second hypervisor may be deployed to address specific tasks or locations in the enterprise. For example, a second hypervisor may be deployed for particular workgroups or branch office locations. Similarly, the second hypervisor may be used for disaster recovery, desktop virtualization support, private cloud computing and so on.
This was first published in April 2013