AZ-305 Designing Microsoft Azure Infrastructure Solutions Exam

Recommend a Recovery Solution for Azure and Hybrid Workloads that Meets Recovery Objectives

Assess Backup and Replication Strategies

Backup and replication strategies are vital for ensuring the reliability and data integrity of applications in Azure. These strategies make recovery swift and cost-effective, particularly when using Azure Site Recovery (ASR) and different SQL Server business continuity and disaster recovery (BCDR) technologies.

SQL Server Disaster Recovery Capabilities

SQL Server offers numerous disaster recovery options to ensure data availability and integrity:

• Failover clustering provides automated switching during hardware or software failures, maintaining application availability.
• Always On availability groups enhance availability by keeping multiple copies of a database, ensuring high uptime.
• Database mirroring ensures database availability via high-performance asynchronous replication.
• Log shipping periodically transfers transaction logs to a secondary system to maintain updates.
• Active geo-replication asynchronously replicates databases across geographical locations for near-real-time failover.
• Auto-failover groups automate the failover process, reducing downtime.

Combining BCDR Technologies with Azure Site Recovery

Azure Site Recovery can be paired with these BCDR technologies to meet specific recovery time objective (RTO) and recovery point objective (RPO) requirements. For example:

• Always On availability group ensures minimal data loss with asynchronous replication, suitable for high-performance environments.
• Failover clustering (Always On Failover Cluster Instances, FCI) uses shared storage to ensure no data loss, although it requires a more complex setup.
• Database mirroring and log shipping are effective for scenarios with higher permissible data lag, offering cost benefits over synchronous options.

Snapshots and Recovery Points

Azure Site Recovery generates snapshots to create recovery points:

• Crash-consistent snapshots are taken every five minutes and capture disk states without in-memory data, suitable for most applications.
• App-consistent snapshots include in-memory data and ongoing transactions, using Volume Shadow Copy Service (VSS) for Windows systems. These are essential to maintain database consistency but are more performance-intensive.

Failover and Failback Processes

Testing failover scenarios is crucial after setting up replication:

1. Failover: Trigger the operation switch to the Azure-based VM during a failure event.
2. Commit failover: Finalize the switch, starting operations on the backup VM.
3. Failback: Prepare the primary site for a switch back from Azure, involving setting up a temporary process server in Azure.

Multi-VM Consistency

For applications needing consistent data across multiple VMs:

• Multi-VM replication groups ensure recovery points are consistent across related VMs by grouping them. However, this may impact performance and should only be enabled when necessary to maintain operational consistency.

In conclusion, selecting the right backup and replication strategy involves balancing RTO/RPO requirements against cost and regulatory constraints. Ensuring optimal configurations tailored to business continuity solutions enhances the robustness and resilience of both cloud-based and hybrid deployments. Using Azure Site Recovery coupled with SQL Server's various BCDR technologies can significantly improve the reliability of your disaster recovery plans.