A company’s API receives real-time data from monitoring devices and stores it in an Amazon RDS DB instance. The incoming data volume fluctuates and during heavy traffic the API times out because the database cannot process the write volume. A solutions architect must minimize database connections and ensure no data loss during heavy traffic. Which solution meets these requirements?

The correct solution uses Amazon SQS as a buffer, decoupling the API from the database. When traffic is high, SQS queues the messages, preventing the API from timing out and ensuring no data loss. A Lambda function then processes messages from the queue at a rate the RDS instance can handle, minimizing direct database connections and preventing overload. Increasing DB instance size might offer temporary relief but doesn't address fluctuating loads efficiently and can be costly. A Multi-AZ configuration provides high availability and disaster recovery but doesn't distribute write load or prevent a single instance from being overwhelmed by a high volume of writes. Using Amazon SNS is less suitable because SNS is a publish/subscribe service primarily for fan-out messaging, not for buffering and reliably