MLA-C01 Practice Question 49: SageMaker Savings Plans
AWS Certified Machine Learning Engineer – Associate (MLA-C01) · Test 5 · question 9 of 10
A company's SageMaker AI spend is steady: real-time endpoints that run around the clock, plus nightly training jobs, together use a consistent baseline of compute every hour. Over the next three years the team expects to move endpoints from ml.c5 to newer instance families, and to move some workloads from us-east-1 to eu-west-1. Finance wants the largest discount on this baseline that does not lock the team into specific instance types or Regions.
Which purchasing option should the team choose?
Show answer and explanation
Answer:
SageMaker AI Savings Plans lower SageMaker AI instance prices in exchange for committing to a consistent amount of usage, measured in dollars per hour, over a one-year or three-year term. The discount applies *regardless of instance family, instance size, component, or AWS Region*, so it still applies after the move to newer families and to eu-west-1. It also covers both endpoint and training usage, which suits a steady baseline (option C).
- A is wrong because EC2 Reserved Instances apply to Amazon EC2 instances, not to SageMaker AI ML instances. Reservations tied to one instance type would also conflict with the planned change of family. - B is wrong because real-time endpoints cannot run on Spot capacity. Managed Spot Training applies to training jobs, and an always-on endpoint that disappears when capacity is reclaimed would not be acceptable anyway. - D is wrong because Compute Savings Plans cover EC2, AWS Fargate, and AWS Lambda usage. They do not apply to SageMaker AI instance usage, which has its own Savings Plan.
Reference: What are Savings Plans?