Skip to content

Backup and Recovery Basics

Protecting your data from loss and corruption is crucial for enterprise workloads on Azure. Azure Backup and Site Recovery provide integrated disaster recovery and backup.

Backup and DR Methods

Method Service SLA (Availability) RPO/RTO Nature Use Case
Snapshots N/A Point-in-time copy; restore time depends on workflow Quick rollback before risky changes
Azure Backup Backup service availability target (for service operation) Depends on policy frequency and restore scope Long-term retention and policy-driven protection
ASR ASR service availability target Replication-based RPO and failover-based RTO Regional/datacenter disaster recovery

SLA vs RPO/RTO

Concept What it Means What it Does Not Mean
SLA Azure service availability commitment (for example, Azure Backup service 99.9% availability) A guarantee that your workload will always meet a specific RPO/RTO
RPO Maximum acceptable data-loss window for your workload A fixed value provided automatically by Azure without your design choices
RTO Maximum acceptable recovery time for your workload A fixed restore duration independent of data size and runbook readiness

Backup Architecture

Azure Backup for Azure VMs uses VM backup extensions with policy-driven scheduling into a Recovery Services vault.

graph TD
    P[Backup Policy] --> J[Scheduled Backup Job]
    J --> V[VM Snapshot + Backup Extension]
    V --> R[Recovery Services vault]
    R --> X[Restore Operations]

Note

RPO (Recovery Point Objective) is the maximum period of data loss, while RTO (Recovery Time Objective) is the time taken to restore services.

Tip

Azure Backup SLA describes the backup service availability, not guaranteed workload recovery outcomes. Validate your expected RPO/RTO with regular restore testing.

Note

A Recovery Services vault stores backup data and can be configured with locally-redundant (LRS), geo-redundant (GRS), or zone-redundant storage (ZRS), depending on scenario and region support.

See Also

Sources