MLS-C01 Question 6
Select 2You are a data scientist setting up a machine learning workflow for a retail company. The company wants to store their historical sales data (structured data), customer reviews (semi-structured JSON files), and product images (unstructured data) in a centralized repository to be used for training ML models. Which AWS services should you use to create this repository while meeting scalability and cost-efficiency requirements?
- A
Amazon S3
- B
Amazon RDS
- C
Amazon DynamoDB
- D
Amazon Redshift
- E
Amazon EFS
Show answer and explanation
Correct answers: A, D
Explanation
Amazon S3 is ideal for storing a variety of data types (structured, semi-structured, and unstructured) cost-effectively, making it a key component of a centralized ML repository. Additionally, Amazon Redshift is well-suited for storing and querying structured data like historical sales data efficiently, complementing S3's capabilities. Together, these services provide a scalable and cost-effective solution for this ML use case. Other options like RDS, DynamoDB, and EFS are either not optimized for ML data repositories or lack the scalability and cost-efficiency required for this scenario.
- A. Correct.
Amazon S3 is a highly scalable and cost-efficient object storage service suitable for storing unstructured data like images, semi-structured data like JSON files, and even structured data in CSV format. It is the best choice for centralizing diverse data types for ML workloads.
- B. Incorrect.
Amazon RDS is a managed relational database service suited for OLTP workloads, and while it can store structured data, it is not ideal for semi-structured or unstructured data. It is also less cost-efficient for large-scale ML repositories compared to Amazon S3.
- C. Incorrect.
Amazon DynamoDB is a NoSQL database service optimized for low-latency and high-throughput workloads involving key-value or document data. It is not suitable for storing large volumes of diverse data types like images or structured data for ML workflows.
- D. Correct.
Amazon Redshift is a fully managed data warehouse service optimized for analytical queries on structured data. It works well for storing structured historical sales data and running complex queries, making it a good choice for this use case.
- E. Incorrect.
Amazon EFS is a scalable file system for Linux-based applications. While it can store files, it is not the most cost-efficient or scalable option for ML workflows involving diverse data types like structured, semi-structured, and unstructured data.