The correct answer is: Set up Amazon RDS Proxy in front of the RDS PostgreSQL database to manage connection pooling.

During periods of high or unpredictable traffic, AWS Lambda can scale quickly and generate a large number of concurrent executions. Each of these may attempt to establish a separate database connection, which can overwhelm traditional RDS databases due to their limited connection pool. RDS Proxy mitigates this by pooling and reusing database connections, allowing thousands of Lambda invocations to share a smaller number of active database connections. This makes it an ideal solution for serverless applications where connection limits are a common bottleneck. Most importantly, it requires minimal to no change in application code, making it the best fit for this scenario.

The option to throttle the concurrency settings of the Lambda functions to limit simultaneous execution could reduce connection pressure but comes at the cost of degraded application performance and user experience. It also doesnt truly solve the root cause of the issueconnection management.

The choice to upgrade the RDS instance class to a larger size to accommodate more database connections might help temporarily, but it increases costs and still does not scale well with sudden spikes or sustained high concurrency. Additionally, it does not resolve the inefficiencies of numerous short-lived connections from Lambda.

The option to switch the relational database to Amazon DynamoDB with auto-scaling capacity suggests a move away from relational databases altogether. While DynamoDB scales well and is fully managed, it would require significant code and schema changes to the existing application. This contradicts the requirement to make the fewest changes possible.

This question is important because it tests your understanding of how serverless compute models like Lambda interact with traditional databases and what tools AWS provides to bridge architectural mismatches. A common trap in these questions is assuming that scaling the database instance solves the problem. In reality, the bottleneck often lies in how connections are managednot raw instance power.

When facing questions about high-concurrency Lambda applications and relational databases, look for hints about connection limits or timeouts. These typically signal that RDS Proxy is the most elegant and cost-effective solution. Always consider minimal code change and operational efficiency when evaluating serverless scenarios.

Full AWS Practitioner Certification Question