Examkin

Question

Difficulty: Very hardDesign Relational Database HA and DR

A global maritime shipping logistics company is designing the disaster recovery architecture for its core vessel dispatch system. The database layer is deployed on an Azure SQL Managed Instance in the North Europe region.

The company has the following requirements for the disaster recovery site in the West Europe region:
- The Recovery Time Objective (RTO) must be less than 10 seconds.
- The Recovery Point Objective (RPO) must be less than 5 seconds.
- Client applications must connect using a single, unchanging connection string that automatically handles failovers.
- Read-heavy reporting queries must be automatically offloaded to the secondary replica.
- All database replication traffic must remain completely private and isolated from the public internet.

Which two components should you include in the relational database design to meet the requirements? (Select TWO.)

  1. Configure an auto-failover group containing the primary and secondary Azure SQL Managed Instances, and route read-write traffic through the primary listener endpoint.Answer
  2. Establish a Global Virtual Network Peering between the virtual networks hosting the primary and secondary Azure SQL Managed Instances.Answer
  3. C
    Configure active geo-replication between the primary and secondary instances, and configure an Azure Application Gateway to manage the redirection of client connections during a failover.
  4. D
    Configure an Azure Traffic Manager profile using performance routing to monitor the database endpoints and automatically update the application connection strings during a failover.

Answer

Configure an auto-failover group containing the primary and secondary Azure SQL Managed Instances, and route read-write traffic through the primary listener endpoint; and establish a Global Virtual Network Peering between the virtual networks hosting the primary and secondary Azure SQL Managed Instances.
The combination of configuring an auto-failover group and establishing Global VNet Peering meets all the requirements. The auto-failover group provides the required RTO and RPO targets, along with a single listener endpoint that routes write operations to the active primary replica without client configuration. Since Azure SQL Managed Instance is deployed within a private subnet of a virtual network, Global VNet Peering connects the primary and secondary networks privately, ensuring that replication traffic remains isolated from the public internet.

Step-by-Step Solution

1
Determine the relational database tier and feature support.
Identify that the workload uses Azure SQL Managed Instance, which does not support active geo-replication but natively supports auto-failover groups for disaster recovery and automated failover.
This establishes the baseline HA/DR technology aligned with the resource SKU.
2
Select the appropriate routing mechanism for client connection persistence.
Incorporate the auto-failover group's read-write listener endpoint in the design.
This guarantees a single connection string for the client applications that persists across failover events without code or configuration updates.
3
Design the network path between the two regional instances.
Select Global VNet Peering to connect the primary virtual network in North Europe with the secondary virtual network in West Europe.
Azure SQL Managed Instance is isolated within a virtual network subnet, so a private network routing path is required to secure replication traffic without exposing endpoints to the public internet.

Key Concept

Cross-region disaster recovery and network routing for Azure SQL Managed Instance using failover groups and Global VNet Peering.
Rate this question

Topics

Design Relational Database HA and DR

More questions from this topic