CloudGuild · Blog · Cheat sheets · Lessons · Certifications
Understanding High Availability Patterns in AWS
Explore a key exam question on AWS high availability patterns and sharpen your understanding for the SAA-C03 exam.
A common mistake in the AWS Certified Solutions Architect - Associate (SAA-C03) exam is misinterpreting high availability requirements. Candidates often focus on scaling within a single region, overlooking the need for resilience across multiple regions.
The question
A company is launching a new application in AWS that requires high availability across multiple regions. What design pattern should the company use to ensure that the application can withstand regional outages?
- A. Deploy the application in a single region with Auto Scaling.
- B. Implement an active-active architecture across multiple regions.
- C. Use AWS Direct Connect for regional failover.
- D. Set up a multi-region read replica for the database.
Think before you scroll
Before making a selection, consider what high availability truly means in the context of regional outages. The design must ensure that if one region goes down, the application remains operational in others.
The answer
The correct option is B. Implement an active-active architecture across multiple regions. This design pattern ensures that the application can continue functioning even if one region experiences an outage, thereby providing high availability.
Why the other options lose
- A. Deploy the application in a single region with Auto Scaling. This option does not meet the requirement for high availability across multiple regions. If the single region fails, the application becomes unavailable.
- C. Use AWS Direct Connect for regional failover. While Direct Connect is useful for establishing dedicated network connections, it does not inherently provide a mechanism for application availability across regions. It is more about connectivity than availability.
- D. Set up a multi-region read replica for the database. This option might improve read performance and availability for the database layer, but it does not address application-level availability. The application itself could still be down if the primary region fails.
The concept behind it
High availability across multiple regions involves deploying an active-active architecture. This means that the application runs in multiple regions simultaneously, allowing it to handle traffic and failover effectively. When one region fails, the other regions continue to serve the application without interruption.
Exam trap to remember
Always consider true high availability as needing to operate across multiple regions, not just within a single region. The active-active architecture is your go-to design pattern for this requirement.