The correct answer is: Modify the API to send incoming data to an Amazon SQS queue. Configure an AWS Lambda function to be triggered by SQS and write the data to the database.

This solution introduces a decoupled, buffer-based architecture. Amazon SQS serves as a highly durable and scalable queue that can absorb bursts of incoming traffic without overwhelming the database. By using a Lambda function to consume messages from the queue and write them to the RDS instance, the number of active connections to the database is reduced, and writes can be throttled to a manageable rate. This also helps eliminate timeout errors during spikes and ensures no data loss as messages are stored durably in SQS until processed.

Why the other options are incorrect:

Increasing the size of the RDS DB instance might temporarily relieve pressure but does not address the root causeconnection saturation and variable traffic. It also increases costs and may not scale well during unpredictable surges.

Modifying the RDS instance to be Multi-AZ only provides high availability and failover protection. It does not increase write throughput, nor does it allow writing to multiple RDS nodes concurrently.

Sending data to Amazon SNS and invoking Lambda from the SNS topic is not ideal for this use case. SNS is designed for pub/sub messaging and broadcast delivery, not message queuing or traffic smoothing. Unlike SQS, SNS does not guarantee message ordering and is not well-suited for queue-based decoupling patterns.

Why this question is important: It evaluates your ability to design a resilient and scalable ingestion pattern by introducing asynchronous processing and decoupling through AWS messaging services. Understanding when to use SQS versus SNS and how to manage database pressure are critical real-world design skills.

Tip: When facing high write volumes or fluctuating traffic, consider buffering with SQS to prevent database overload. SQS + Lambda is a cost-efficient and scalable pattern for smoothing workloads.

Full AWS Practitioner Certification Question