AZ-305 Question 71
Select 2You manage an e-commerce platform that hosts its primary transactional database on Azure SQL Database (currently in the Standard tier). During major sales events, read-heavy queries for reporting and analytics cause noticeable latency. Management wants to scale read traffic with minimal code changes and minimal downtime. Which two solutions would best address these requirements?
- A
Switch the database to a Business Critical tier and enable read scale-out replicas
- B
Migrate the database to the Hyperscale tier to use built-in read replicas
- C
Deploy Azure Load Balancer in front of the existing single Azure SQL Database instance
- D
Introduce manual sharding at the application level to distribute read workloads
- E
Upgrade to a Premium Elastic Pool to share resources among multiple databases
Show answer and explanation
Correct answers: A, B
Explanation
The Business Critical service tier and Hyperscale tier both offer read scale-out capabilities for Azure SQL Database, enabling the creation of multiple replicas to handle read-intensive workloads. This approach meets the criteria of minimal code changes and downtime. For more detailed guidance, refer to the official Azure SQL Database documentation at https://learn.microsoft.com/azure/azure-sql/database/ .
- A. Correct.
Option 1 is correct. The Business Critical service tier supports read scale-out by automatically provisioning read-only replicas. This allows you to offload read workloads with minimal code changes, since your application can point read queries to the assigned read replica endpoint.
- B. Correct.
Option 2 is correct. The Hyperscale service tier offers multiple read replicas for handling intensive read workloads. It also allows near-instant scale-up without major architectural changes or downtime, making it an ideal choice when you need large-scale read performance.
- C. Incorrect.
Option 3 is incorrect. Azure Load Balancer is designed for distributing network traffic to multiple compute endpoints (e.g., VMs). It cannot split read traffic on a single Azure SQL Database instance, and does not inherently provide additional database replicas for scale-out.
- D. Incorrect.
Option 4 is incorrect. While manual sharding can spread workloads across multiple databases, it typically requires significant application code changes and complex data distribution logic. This does not meet the goal of 'minimal code changes.'
- E. Incorrect.
Option 5 is incorrect. Premium Elastic Pools share compute resources across multiple databases but do not inherently provide additional read replicas. They are intended for multi-database workloads rather than scaling reads on a single database.