Examkin

Question

Difficulty: MediumDesign Azure Site Recovery Solutions

A logistics company hosts a legacy portal on-premises on VMware. The application tier consists of three stateless web virtual machines (VMs) with minimal write activity. The data tier consists of a single high-transaction SQL Server VM. The SQL Server VM has a system disk, a data disk averaging 8 MB/s8 \text{ MB/s} write churn, and a transaction log disk averaging 32 MB/s32 \text{ MB/s} write churn. The company requires a disaster recovery (DR) solution to Azure with a Recovery Point Objective (RPO) of under 10 seconds for the database and under 4 hours for the web tier.

Which disaster recovery design should you recommend?

  1. Replicate the application VMs using Azure Site Recovery, and replicate the database using SQL Server Always On Availability Groups to SQL Server on Azure VMs.Answer
  2. B
    Replicate all VMs, including the SQL Server VM, to Azure Premium SSDs using Azure Site Recovery.
  3. C
    Replicate all VMs to Azure using Azure Site Recovery, and configure Read-Access Geo-Redundant Storage (RA-GRS) to support active write operations on the database replica disks.
  4. D
    Replicate both the application VMs and the SQL Server VM using Azure Backup with instant restore points configured every 15 minutes.

Answer

Replicate the application VMs using Azure Site Recovery, and replicate the database using SQL Server Always On Availability Groups to SQL Server on Azure VMs.
The correct design replicates the stateless application VMs using Azure Site Recovery and utilizes SQL Server Always On Availability Groups to replicate the database. The database's transaction log disk churn of 32 MB/s32 \text{ MB/s} exceeds the 25 MB/s25 \text{ MB/s} limit per Premium SSD disk for Azure Site Recovery VMware-to-Azure replication. Replicating the database natively via Always On Availability Groups bypasses this limit and meets the low RPO of under 10 seconds.

Step-by-Step Solution

1
Evaluate the database write churn requirements against Azure Site Recovery limits.
The SQL Server VM has a transaction log disk experiencing 32 MB/s32 \text{ MB/s} write churn, which exceeds the Azure Site Recovery limit of 25 MB/s25 \text{ MB/s} per Premium SSD disk for VMware-to-Azure replication.
Identifying capacity limits prevents replication failures and unhealthy replication states in Azure Site Recovery.
2
Evaluate the Recovery Point Objective (RPO) requirements for both tiers.
The web tier requires an RPO under 4 hours, which is easily met by Azure Site Recovery. The database tier requires an RPO of under 10 seconds, which Azure Site Recovery cannot guarantee.
Choosing the replication technology must align with the target business continuity and disaster recovery metrics.
3
Determine the optimal replication strategy for the hybrid architecture.
Combine Azure Site Recovery for the application tier with native SQL Server Always On Availability Groups (configured for asynchronous or synchronous commit) for the database tier.
This hybrid approach ensures high-churn databases are replicated via application-native mechanisms to satisfy both churn limits and low RPO requirements, while web VMs leverage standard VM replication.

Key Concept

Designing disaster recovery solutions by aligning workload write churn metrics and RPO requirements with Azure Site Recovery support limits and native replication alternatives.
Rate this question

Topics

Design Azure Site Recovery Solutions

More questions from this topic