A solutions architect is designing a new service behind Amazon API Gateway. Request patterns will be unpredictable and can spike suddenly from 0 to over 500 requests per second. The total persisted data is currently less than 1 GB with uncertain future growth and can be queried with simple key-value requests. Which combination of AWS services would meet these requirements? (Choose two.)

AWS Lambda is ideal for unpredictable, spiky workloads because it's serverless and scales automatically from zero to meet demand without requiring pre-provisioning. This perfectly addresses the requirement for sudden spikes from 0 to over 500 requests per second. Amazon DynamoDB is a fully managed NoSQL database that offers single-digit millisecond performance at any scale, making it suitable for simple key-value requests and handling uncertain future data growth. Its serverless nature and on-demand capacity mode align well with unpredictable access patterns. AWS Fargate and Amazon EC2 Auto Scaling involve managing containers or EC2 instances, which would incur costs even when idle and require more operational overhead to manage scaling, unlike Lambda. MySQL-compatible Amazon Aurora is a relational database, which is overkill for simple key-value requests and generally has higher operational overhead and less flexible scaling for spiky, serverless workloads compared to DynamoDB.