SAA-C03 Question 339
Select 2A retail company needs to ingest and transform real-time streaming data from multiple sources, such as website clickstreams, IoT device metrics, and mobile app interactions. The company wants to ensure low latency, reliability, and the ability to run SQL-based transformations on the incoming data. Which combination of AWS services would best meet this requirement?
- A
Amazon Kinesis Data Streams and AWS Lambda
- B
Amazon Kinesis Data Analytics and Amazon Kinesis Data Streams
- C
Amazon S3 and AWS Glue
- D
Amazon DynamoDB and Amazon EMR
- E
Amazon Redshift and Amazon Data Pipeline
Show answer and explanation
Correct answers: A, B
Explanation
The combination of Amazon Kinesis Data Streams and Amazon Kinesis Data Analytics is the best solution for this scenario. Kinesis Data Streams enables the ingestion of high-throughput, real-time data from multiple sources, and Kinesis Data Analytics allows you to run SQL-based transformations on the streaming data. This setup ensures low latency, reliability, and the ability to handle real-time transformations, meeting the company's requirements.
- A. Correct.
Amazon Kinesis Data Streams and AWS Lambda can work together to ingest real-time data with low latency. Kinesis Data Streams allows you to process high-throughput data streams, and AWS Lambda can be used for lightweight transformations. However, this solution alone does not include SQL-based transformation capabilities.
- B. Correct.
Amazon Kinesis Data Analytics and Amazon Kinesis Data Streams is the best choice for this scenario. Kinesis Data Streams handles real-time data ingestion, while Kinesis Data Analytics provides the ability to run SQL-based transformations on the streaming data.
- C. Incorrect.
Amazon S3 and AWS Glue are better suited for batch processing rather than low-latency, real-time streaming use cases. S3 serves as a storage layer, and AWS Glue is typically used for ETL tasks on stored data.
- D. Incorrect.
Amazon DynamoDB and Amazon EMR are not ideal for real-time streaming and SQL-based transformations. DynamoDB is a NoSQL database and not a data ingestion service, while EMR is primarily used for big data processing with frameworks like Hadoop and Spark.
- E. Incorrect.
Amazon Redshift and Amazon Data Pipeline are better suited for data warehousing and batch processing workflows rather than real-time ingestion and transformation.