Operate a Multicloud Environment
Multicloud items test whether you understand which Nutanix component runs where, and how a single operating model extends beyond the datacenter.
The pieces
- Prism Central is the control plane: categories, projects, policies, Calm automation and Flow network security apply consistently across registered clusters, wherever they run.
- Nutanix Cloud Clusters (NC2) runs a real Nutanix cluster on AWS or Azure bare metal, so AHV, AOS and Prism behave the same way in the cloud.
- Nutanix Move migrates VMs from VMware, Hyper-V, AWS or Azure into AHV with minimal downtime.
- Xi Leap provides cloud-based DR; Frame delivers desktops as a service; Buckets provides S3-compatible object storage; Files provides scale-out NFS/SMB.
Operating rules
- Prism Central must be equal to or newer than every registered Prism Element.
- Categories created in Prism Central are what multicloud policies select on - design the taxonomy before building policies.
- Data gravity and egress cost decide placement as much as latency does.
Worked example
A team wants on-prem VMs and an AWS footprint managed from one pane. Deploy an NC2 cluster in AWS, register it to the existing Prism Central, apply the same categories and protection policy, then use Move to migrate the chosen VMs. Applications stay in one inventory and the protection policy now covers both sites.
Traps
- Assuming Prism Central can manage a cluster running a newer AOS than itself.
- Designing multicloud policy around clusters rather than categories.
- Ignoring cross-cloud latency and egress for chatty applications.
- Treating cloud instances as automatically highly available - NC2 clusters still need correct RF and FT sizing.