I85PREP
⌘ K
Nutanix NCP-MCI study materials
Nutanix NCP-MCI study resources

Nutanix NCP-MCI: Protect and Recover Data

Nutanix NCP-MCI · Work through your exam, one topic at a time.

All guide sections

Protect and Recover Data

Data protection items test matching an RPO to the right Nutanix mechanism and understanding what a snapshot actually protects you from.

Protection Domains

  • A Protection Domain (PD) groups VMs and consistency groups, applies snapshot schedules (frequency, retention), optionally replicates to a remote site, and can be mapped to a DR runbook in Prism Central.
  • A consistency group keeps multi-VM application writes consistent across snapshots.

Choosing the mechanism

  • Local snapshots: redirect-on-write, space-efficient, instantaneous - but on the same cluster, so not DR.
  • Async replication (PD): minutes-to-hours RPO, bandwidth-friendly, works over WAN.
  • NearSync: RPO of seconds (configured 1–15 minutes) using frequent snapshots plus the oplog, within latency limits.
  • Metro Availability: synchronous stretch cluster, RPO 0, requires low-latency links and stretched VLANs; a witness handles split-brain.

Worked example

An application demands zero data loss between two datacenters 5 km apart with dark fibre. That maps to Metro Availability, not NearSync - NearSync gives an RPO in the seconds-to-minutes range. You then verify the VLANs are stretched so VMs keep their addresses after failover, and that a witness is deployed.

Traps

  • Promising RPO 0 with async PD.
  • Forgetting the remote site needs a compatible container and AOS version.
  • Metro over high-latency links - it will not stay in sync.
  • No DR test: a runbook that has never been executed is an assumption.