Nutanix Enterprise Cloud Fundamentals
What this domain really tests
Nutanix is a hyperconverged platform: AOS (the distributed cluster software) runs on every node alongside a standard hypervisor - AHV, ESXi, or Hyper-V - with one Controller VM (CVM) per node owning that node's local disks. Everything else (Prism, Flow, Files, Objects, Volumes, Calm, Leap, Era, Move, Karbon) is a service layered on that foundation.
Core concepts: HCI replaces the SAN/three-tier stack; scale-out is linear (add a node, add CPU, memory, and storage together); data locality means reads are served by the local CVM and writes land locally before being replicated; the platform is software-defined and upgrades are rolling and non-disruptive.
Traps
- AOS is not the hypervisor. AHV is the hypervisor; AOS is storage + cluster services. Questions like "which component runs the VMs?" want AHV.
- Prism Element manages a single cluster; Prism Central manages many clusters and unlocks categories, RBAC, protection policies, Calm, Leap, and Flow security.
- Product matching is heavily tested: Flow = microsegmentation/network security, Files = SMB/NFS file services, Objects = S3-compatible object store, Volumes = block/iSCSI for external hosts, Leap = DR orchestration, Move = migration from other platforms, Era = database lifecycle, Karbon = Kubernetes.
- "AHV is mandatory" is false - the CVM also runs on ESXi and Hyper-V.
Rule of thumb
If a question asks where storage I/O is served, the answer is the CVM on the same host. Stop or remove that CVM and the host loses its storage path - that single idea explains data locality, I/O paths, and why AOS upgrades are done one CVM at a time.