MLS-C01 · Question #170
A machine learning specialist stores IoT soil sensor data in Amazon DynamoDB table and stores weather event data as JSON files in Amazon S3. The dataset in DynamoDB is 10 GB in size and the dataset…
The correct answer is D. Crawl the data using AWS Glue crawlers. AWS Glue provides the least administrative overhead for this ETL scenario. Glue crawlers can automatically discover and catalog data from both Amazon DynamoDB and Amazon S3, inferring schemas and populating the AWS Glue Data Catalog. Glue ETL jobs can then join, transform, and…
Question
A machine learning specialist stores IoT soil sensor data in Amazon DynamoDB table and stores weather event data as JSON files in Amazon S3. The dataset in DynamoDB is 10 GB in size and the dataset in Amazon S3 is 5 GB in size. The specialist wants to train a model on this data to help predict soil moisture levels as a function of weather events using Amazon SageMaker. Which solution will accomplish the necessary transformation to train the Amazon SageMaker model with the LEAST amount of administrative overhead?
Options
- ALaunch an Amazon EMR cluster.
- BCrawl the data using AWS Glue crawlers.
- CEnable Amazon DynamoDB Streams on the sensor table.
- DCrawl the data using AWS Glue crawlers.
How the community answered
(29 responses)- A10% (3)
- B14% (4)
- C3% (1)
- D72% (21)
Explanation
AWS Glue provides the least administrative overhead for this ETL scenario. Glue crawlers can automatically discover and catalog data from both Amazon DynamoDB and Amazon S3, inferring schemas and populating the AWS Glue Data Catalog. Glue ETL jobs can then join, transform, and output the combined dataset in a SageMaker-compatible format (e.g., CSV or RecordIO in S3), all without managing servers. Options B and D both state 'Crawl the data using AWS Glue crawlers' - D is the correct answer. Launching an EMR cluster (A) requires cluster configuration, management, and writing Spark/Hive jobs - significantly more administrative overhead. Enabling DynamoDB Streams (C) captures real-time incremental changes and is designed for event-driven pipelines, not for a one-time or batch training data extraction use case.
Topics
Community Discussion
No community discussion yet for this question.