An online food company uses an API Gateway HTTP API integrated with a Lambda that writes orders to DynamoDB. New partners may require additional Lambda functions to receive orders. The company has an S3 bucket and wants every order and update stored in S3 for analysis with minimal development effort. What is the simplest way to ensure all orders and updates are written to S3?

Enabling DynamoDB Streams on the orders table and triggering a Lambda function from the stream is the simplest and most efficient solution. DynamoDB Streams capture every item-level modification (insertions, updates, and deletions) in near real-time. The Lambda function can then process these stream records and write them to S3, ensuring all orders and updates are captured without modifying the existing order processing logic. Creating a new Lambda and API endpoint (option 1) adds unnecessary complexity and latency, requiring the original Lambda to make an additional API call. Modifying the existing Lambda to publish to Kinesis (option 2) or SNS (option 4) would require significant changes to the core order processing logic, increasing development effort and potential for errors. DynamoDB Streams provide a native, event-driven way to react to changes in the table.