CloudGuild · Blog · Cheat sheets · Lessons · Certifications
Amazon Aurora Serverless & Global Database — Scalable, on-demand database solutions: when to use what (and how the exam tests it)
Learn when to use Amazon Aurora Serverless vs. Global Database and how the AWS Certified Solutions Architect - Associate exam tests your knowledge on these solutions.
Amazon Aurora, compatible with MySQL and PostgreSQL, is a key player in AWS's database offerings. It offers two powerful features: Aurora Serverless and Global Database. Understanding when to use each can significantly impact application performance and cost.
When to Choose Amazon Aurora Serverless
Aurora Serverless is ideal for workloads that are variable, unpredictable, or infrequent. This includes scenarios like development and test environments where database usage fluctuates. The automatic scaling is a critical feature, allowing it to adjust compute capacity based on demand. However, it’s important to note that scaling can take time, making it unsuitable for real-time applications.
When to Choose Global Database
Global Database is designed for applications that require low-latency access across multiple geographical regions. If you have users distributed globally, this option enables fast read and write operations in various AWS Regions. It’s particularly beneficial for applications needing high availability and rapid failover capabilities.
Key Architectural Decisions
When deciding between these two options, consider the following:
- Database Compatibility: Choose between MySQL or PostgreSQL based on your application ecosystem.
- Capacity Settings for Aurora Serverless: Define the minimum and maximum Aurora Capacity Units (ACUs) to optimize cost and performance.
- Replication Strategy for Global Database: Determine the read/write regions and understand the limitations, such as the maximum of five read replicas.
| Feature | Aurora Serverless | Global Database |
|---|---|---|
| Scalability | Auto-scales based on demand | Fixed capacity, supports replication |
| Use Case | Variable workloads, dev/test | Low-latency access across regions |
| Latency | May introduce delay during scaling | Low latency for reads/writes |
| Read Replicas | Not applicable | Up to 5 in different regions |
Gotchas & Exam Traps
Be aware of the following when preparing for the exam:
- Scaling Limitations: Aurora Serverless may take time to scale, which can impact performance in real-time applications. This is a common trap in exam questions.
- Replication Constraints: Global Database allows only five read replicas across regions. Questions may test your understanding of how this affects latency and availability.
How the Exam Tests This
Exam questions often focus on:
- Scenario-based questions: Expect questions that present a situation requiring you to choose between Aurora Serverless and Global Database based on specific requirements.
- Limitations and Trade-offs: Questions may ask about the constraints of each option, such as scaling delays or the maximum number of replicas.
- Best Practices: You may encounter questions that test your knowledge of best practices for implementing these solutions effectively.
The Rule to Remember
Aurora Serverless excels in variable workloads, while Global Database is best for low-latency, geographically distributed applications.
With this understanding, you can make informed decisions in real-world scenarios and on the exam.