Data Protection, Backup, and Disaster Recovery
This domain tests the difference between a snapshot, a replica, and a backup, and how RPO and RTO select the tool.
Snapshots and protection domains
- Local snapshots are space-efficient and crash-consistent by default; application consistency requires NGT/VSS.
- Protection domains group VMs; consistency groups keep a multi-VM application mutually consistent. Schedules set RPO and retention sets how far back you can restore.
Replication and DR
- Async replication to a remote site covers RPO measured in hours and tolerates WAN latency.
- NearSync drives RPO down to minutes and relies on a witness for automatic failover in two-site designs.
- Leap orchestrates recovery plans and runbooks, including non-disruptive tests using cloned or isolated networks.
- Files, Volumes, and Objects each have their own replication mechanics; third-party backup integrates through snapshot and VSS paths.
Worked example
Tier 1 needs RPO of five minutes and RTO of thirty: NearSync plus a witness plus a tested Leap runbook. Tier 2 tolerates four hours: nightly async replication suffices. Tier 3 only needs point-in-time recovery from accidental deletion: that is a backup requirement, and a replica will not save you, because deletions replicate too.
Traps
Treating a snapshot as a backup (same cluster, same failure domain); assuming replication protects against logical corruption; omitting consistency groups for multi-VM apps; running a two-site design without a witness and expecting automatic failover.