MLA-C01 Practice Question 91: Intelligent Tiering Training Data
AWS Certified Machine Learning Engineer – Associate (MLA-C01) · Test 10 · question 1 of 10
A company stores 300 TB of historical training datasets in Amazon S3 Standard. Access is unpredictable: a dataset can go untouched for months, then be read in full when a team retrains a model on short notice. The company wants storage costs to fall automatically for data that is not being used, with no retrieval fees, no delay when a dataset is suddenly needed, and no lifecycle rules to design and tune.
Which storage class should the company use?
Show answer and explanation
Answer:
S3 Intelligent-Tiering is built for data with unknown, changing, or unpredictable access patterns. For a small monthly monitoring and automation fee per object, it moves objects to lower-cost tiers when they are not accessed and back to the Frequent Access tier when they are, with no performance impact and no retrieval fees. Its automatic access tiers serve data in milliseconds, so a sudden retraining run reads the dataset without delay or a restore step (option D).
- A is wrong because Standard-IA charges a per-GB retrieval fee. Reading an entire dataset for an unplanned retraining can wipe out the savings, and a lifecycle rule based on age does not know which datasets are actually in use. - B is wrong because objects in Glacier Flexible Retrieval must be restored before they can be read, which takes minutes to hours and costs extra, the delay the company ruled out. - C is wrong because One Zone-IA also charges retrieval fees, and it keeps data in a single Availability Zone, lowering resilience for years of training history.
Reference: Understanding and managing Amazon S3 storage classes