Which AWS service can be used to process streaming data with serverless compute and is commonly integrated with Kinesis for real-time analytics?

• A. Amazon EC2
• B. AWS Lambda
• C. Amazon EMR
• D. AWS Fargate

Answer: (B)

Processing streaming data with serverless compute that works well for real-time analytics is about reacting to data as it arrives without managing servers yourself. The service that fits this pattern best is a serverless, event-driven compute option that can be configured to consume records directly from a streaming source like Kinesis. When you set up this integration, a function runs in response to new data in the stream, processing batches of records, performing transforms or enrichment, and then routing results to destinations such as data lakes or analytics stores. The strengths here are automatic scaling in response to the stream’s throughput, billing only for actual compute time used, and minimal maintenance since there’s no server provisioning.

This approach is particularly effective with Kinesis because you can have a dedicated function process each shard’s data in real time, enabling low-latency analytics and transformations as data flows in. The function can be kept simple and stateless, and you can edge-case manage failures, retries, and idempotency within the processing logic.

Other options would require managing servers or clusters or are not as tightly aligned with the event-driven, on-demand model for streaming data: a traditional virtual machine-based approach incurs operational overhead; a managed analytics cluster focuses more on batch or long-running workloads; and container-based serverless options exist but don’t integrate as naturally with Kinesis for real-time stream processing.