4.1.1. Design for Azure Backup
💡 First Principle: A reliable, automated, and secure backup solution is the fundamental safeguard against data loss from accidental deletion, corruption, or ransomware, ensuring data can be recovered to a known good state.
Scenario: You are designing a backup solution for a critical production application that includes Azure Virtual Machines and Azure SQL Databases. You need to ensure daily backups are taken, retained for 90 days, and then moved to a long-term archival tier for 7 years to meet compliance.
Azure Backup is a fully managed, cloud-native backup solution that protects your data across Azure workloads and on-premises environments.
Key Design Considerations:
- Backup Targets: Azure Backup supports a wide range of targets, including Azure VMs, SQL Server (in Azure or on-premises), Azure Files, Azure Blobs, and physical or virtual on-premises servers.
- Recovery Services Vault: Centralizes backup management, policy configuration, and storage, simplifying monitoring and compliance across all protected assets.
- Retention Policies: Flexible short-term and long-term retention options allow organizations to meet regulatory and business requirements for data recovery.
- Cost Optimization: Operates on a pay-as-you-go model with tiered storage (e.g., moving older backups to cheaper Azure Blob storage Archive tier), enabling efficient cost management.
- Security: Provides encryption at rest and in transit, supports immutability to prevent tampering or ransomware attacks, and integrates with Azure security controls.
Design decisions in practice:
| Decision | Guidance |
|---|---|
| Vault redundancy | Chosen at creation and hard to change: LRS for non-production, GRS for anything that must survive the loss of the region, ZRS where in-region zone resilience is enough |
| Policy | Frequency sets the recovery point objective; retention sets how far back you can go. Split policies by workload class rather than applying one maximum retention to everything |
| Tamper protection | Immutability stops retention being shortened or recovery points deleted before expiry. Soft delete recovers a deleted item within a window. Multi-user authorization adds an approval step to destructive operations |
| What it protects | Azure VMs, SQL Server and SAP HANA IN VMs, Azure Files, Blobs, and on-premises servers through the MARS agent |
Backup answers a different threat from replication, and questions rely on the difference. Replication faithfully copies whatever happened, including a deletion or an encryption event — so it protects against infrastructure loss and not against mistakes or malice. Backup keeps earlier states, which is why the answer to "a user deleted a file eleven days ago" and to ransomware is always backup, never geo-redundancy.
⚠️ Exam Trap: Azure SQL Database (the PaaS service) is not protected by a Recovery Services vault. It has its own automated backups and long-term retention configured on the database. The vault covers SQL Server running inside a VM.
⚠️ Common Pitfall: Not testing backups. An untested backup is not a reliable recovery strategy. Regularly perform test restores to a non-production environment to validate data integrity and the recovery process.
Key Trade-Offs:
- Retention Period vs. Cost: Longer retention periods and more frequent backups provide more recovery points but increase storage costs. The policy must balance business recovery needs with budget.
Reflection Question: How does designing for Azure Backup, leveraging its wide range of backup targets, centralized Recovery Services Vault, and flexible retention policies, fundamentally ensure reliable, scalable, and cost-effective data protection and recovery for workloads across cloud and hybrid environments, ensuring business continuity?
