Sensors mounted across factories stream environmental data (humidity, light, etc.) to AWS. The company needs real-time streaming and analysis, and must immediately notify the operations team when any parameter falls outside acceptable ranges. Which design meets these requirements?

The correct design uses Amazon Kinesis Data Streams for real-time data ingestion, as it's designed for high-throughput, continuous data streams. An AWS Lambda function can directly consume and process records from a Kinesis Data Stream in real time, performing the necessary analysis and anomaly detection. Amazon SNS is the appropriate service for sending immediate, fan-out notifications to a team. Incorrect options: Kinesis Data Firehose is primarily for loading streaming data into data lakes, data stores, and analytics services, not for direct real-time analysis and alerting by Step Functions. Step Functions are better suited for orchestrating workflows, not consuming raw streaming data directly. Amazon MSK is a good streaming service, but configuring a trigger in MSK to start a Fargate task for every data analysis event is less efficient and more complex than Lambda's direct integration with Kinesis Data Streams for real-time processing. SES is for email, while SNS is better for general notifications to various endpoints. Kinesis Data Analytics is suitable for real-time analytics, but pairing it with a separate ECS service to consume and analyze results adds unnecessary complexity. Kinesis Data Analytics can often perform the analysis itself or integrate with other services more directly. Again, SES is not the primary notification service for operational alerts.