MLS-C01 Question 16
Single answerA data science team is building a machine learning pipeline to train models using large datasets. The datasets need to be ingested and stored in a cost-effective, scalable, and highly durable storage solution. Additionally, the data will be accessed in a read-heavy manner during training, but frequent updates are not required. Which storage medium should they choose?
- A
Amazon S3
- B
Amazon Elastic File System (Amazon EFS)
- C
Amazon Elastic Block Store (Amazon EBS)
- D
Amazon RDS
Show answer and explanation
Correct answer: A
Explanation
Amazon S3 is the most appropriate choice for this scenario because it provides a highly durable, scalable, and cost-effective object storage solution. It is optimized for read-heavy workloads and is commonly used in machine learning pipelines to store large datasets. Other storage options like Amazon EFS, Amazon EBS, and Amazon RDS are better suited for specific use cases such as shared file systems, low-latency block storage, and relational database management, respectively.
- A. Correct.
Amazon S3 is a cost-effective, scalable, and highly durable object storage service suitable for storing large datasets. It supports high read throughput and is ideal for machine learning pipelines where frequent updates are not required.
- B. Incorrect.
Amazon Elastic File System (Amazon EFS) is a file storage service that supports concurrent access by multiple instances. However, it is more suited for shared file systems and use cases requiring frequent updates, not for cost-effectively storing large datasets accessed in a read-heavy manner.
- C. Incorrect.
Amazon Elastic Block Store (Amazon EBS) is a block storage service optimized for low-latency and high-performance use cases like databases. It is not ideal for storing large datasets in a cost-effective and scalable way.
- D. Incorrect.
Amazon RDS is a managed relational database service. While it is good for structured data and transactional workloads, it is not designed for storing large datasets for machine learning pipelines.