MLS-C01 Question 40
Single answerYou are a Machine Learning Engineer working for a retail company that processes large amounts of customer transaction data to train recommendation models. The data is semi-structured and stored in Amazon S3. You need to preprocess this data efficiently and scale resources dynamically based on the workload. Which solution would best meet these requirements?
- A
Use Amazon EMR with Apache Spark for distributed data processing and dynamic scaling.
- B
Use Amazon RDS with a read replica for processing the transaction data.
- C
Use AWS Glue for preprocessing the data and Amazon EC2 Auto Scaling for scaling resources.
- D
Use Amazon Redshift for data preprocessing and scaling compute resources.
Show answer and explanation
Correct answer: A
Explanation
Amazon EMR is a managed service that allows you to process and analyze large datasets using distributed frameworks like Apache Spark. It is well-suited for preprocessing semi-structured data stored in Amazon S3, and its ability to dynamically scale resources makes it ideal for handling varying workloads efficiently. Other options, such as RDS, Glue, and Redshift, are not optimized for this specific use case.
- A. Correct.
This is the correct answer. Amazon EMR can process large-scale data efficiently using Apache Spark for distributed computing. EMR also supports dynamic scaling to adjust resources based on workload.
- B. Incorrect.
This is incorrect. Amazon RDS is designed for relational database workloads and is not optimized for large-scale distributed data processing or dynamic scaling for big data use cases.
- C. Incorrect.
This is incorrect. While AWS Glue is useful for ETL tasks, it does not provide the same level of flexibility and scalability for distributed data processing as Amazon EMR. EC2 Auto Scaling is also not an optimal solution for dynamic scaling in this scenario.
- D. Incorrect.
This is incorrect. Amazon Redshift is a data warehouse designed for analytical queries on structured data. It is not ideal for semi-structured data preprocessing or dynamic scaling for distributed computing.