Implementing Disaster Recovery
Core concepts
Disaster recovery covers backup, restore, replication, and recovery objectives. Know Windows Server Backup, Azure Backup with the MARS agent and Recovery Services vault, and Azure Site Recovery for replication. Understand RPO (acceptable data loss) and RTO (acceptable downtime). Know Hyper-V Replica, Storage Replica, and backup of domain controllers (authoritative vs non-authoritative restore, DSRM). Understand Bare Metal Recovery, system state backup, and the difference between full, incremental, and differential backups. Also know Azure Backup retention, soft delete, and immutable vaults.
Traps to watch
System state backup includes AD, registry, SYSVOL, and COM+ but not user data. Restoring a domain controller requires DSRM and careful authoritative restore. Azure Backup MARS agent backs up files and system state, not Hyper-V VMs; use the Azure Backup extension or ASR for VMs. Site Recovery is for replication and failover, not long-term retention. Soft delete protects against accidental deletion but is not a backup substitute. Verify RPO/RTO mapping: Storage Replica synchronous gives near-zero RPO but latency limits.
Worked example
A business requires 15-minute RPO and 1-hour RTO for a Hyper-V VM. Rule of thumb: use Hyper-V Replica with 15-minute replication frequency plus tested failover, or Azure Site Recovery for cloud failover; Windows Server Backup cannot meet a 15-minute RPO. Document and test the restore.
Exam approach
Translate requirements into RPO/RTO first, then select the technology. Always ask whether the scenario needs retention, replication, or fast failover.