Protecting On-Premises Workloads with Azure Site Recovery: Failover Strategy Recommendation

Failover Strategy for Protecting On-Premises Workloads with Azure Site Recovery

Question

You plan to use Azure Site Recovery to protect several on-premises physical server workloads. Each server workload is independent of the other. The workloads are stateless.

You need to recommend a failover strategy to ensure that if the on-premises data center fails, the workloads are available in Azure as quickly as possible.

Which failover strategy should you include in the recommendation?

Answers

Explanations

Click on the arrows to vote for the correct answer

A. B. C. D.

D

https://docs.microsoft.com/en-us/azure/site-recovery/site-recovery-failover

To ensure that on-premises physical server workloads are available in Azure as quickly as possible in case of a data center failure, we need to recommend a failover strategy using Azure Site Recovery.

Azure Site Recovery provides different types of replication for different types of workloads. These include:

  1. Asynchronous replication: It replicates data asynchronously, meaning there may be some data loss in case of a failover. It is ideal for non-critical workloads or workloads with a low Recovery Point Objective (RPO).

  2. Synchronous replication: It replicates data synchronously, ensuring zero data loss. It is ideal for critical workloads or workloads with a low Recovery Time Objective (RTO).

As the workloads are stateless and independent of each other, we can choose any replication type. However, to ensure availability as quickly as possible, we need to choose a replication type with a low RTO.

The failover strategy that we should recommend is "Latest app-consistent". This ensures that the latest recovery point is used and ensures application consistency. In case of a failover, the application will be available in Azure with minimal data loss and in a consistent state.

Option A, "Latest", does not ensure application consistency and may result in data loss.

Option C, "Latest multi-VM processed", is used for multi-VM application consistency and is not required for stateless workloads.

Option D, "Latest processed", is not suitable for stateless workloads as it does not ensure application consistency.

Therefore, the recommended failover strategy is "Latest app-consistent".