DEA-C01 Question 276
Select 2You are designing a data pipeline for a retail company that processes data from multiple sources. The data includes structured transactional records, semi-structured clickstream logs, and unstructured product images. Which combination of AWS services should you use to efficiently store and process the data for analytics?
- A
Amazon RDS for structured data, Amazon S3 for semi-structured data, and Amazon Rekognition for unstructured data
- B
Amazon Redshift for structured data, Amazon DynamoDB for semi-structured data, and Amazon EMR for unstructured data
- C
Amazon Aurora for structured data, Amazon S3 for semi-structured data, and Amazon S3 with metadata tagging for unstructured data
- D
Amazon Redshift for structured data, Amazon S3 for semi-structured data, and Amazon Rekognition for unstructured data
- E
Amazon DynamoDB for structured data, Amazon OpenSearch Service for semi-structured data, and Amazon ElastiCache for unstructured data
Show answer and explanation
Correct answers: A, D
Explanation
When modeling structured, semi-structured, and unstructured data on AWS, it is important to use services designed for the specific data types and requirements. Amazon RDS or Amazon Redshift are appropriate for structured data. Semi-structured data benefits from Amazon S3's ability to handle diverse formats at scale. For unstructured image data, Amazon Rekognition provides advanced capabilities for image processing and analysis. Combining these services ensures an efficient and scalable data pipeline tailored to the needs of the retail company.
- A. Correct.
Correct: Amazon RDS efficiently handles structured data like transactional records. Amazon S3 is ideal for storing semi-structured data due to its scalability and support for different formats like JSON or Parquet. Amazon Rekognition is a specialized service for analyzing unstructured images, making it a suitable choice for product image processing.
- B. Incorrect.
Incorrect: Amazon DynamoDB is a NoSQL database, which is not optimal for structured transactional data. Amazon EMR is typically used for big data processing and analytics, but it is not specifically optimized for unstructured image data.
- C. Incorrect.
Incorrect: While Amazon Aurora is suitable for structured data, using Amazon S3 with metadata tagging for unstructured data is not a recommended approach for image analysis. Amazon Rekognition would be more appropriate for this purpose.
- D. Correct.
Correct: Amazon Redshift is designed for structured data warehouses and analytics. Amazon S3 can store semi-structured data in formats like JSON or Avro. Amazon Rekognition is a natural fit for unstructured data like images, enabling advanced image recognition and analysis.
- E. Incorrect.
Incorrect: Amazon DynamoDB is not ideal for structured transactional records. Amazon OpenSearch Service (formerly Elasticsearch) is good for search and analytics on semi-structured data but does not fit this scenario. Amazon ElastiCache is a caching service and not suitable for handling unstructured data like images.