AWS Certified Machine Learning – SpecialtyMachine Learning Implementation and OperationsEasy
A data science team is deploying a new machine learning model for real-time inference on Amazon SageMaker. They observe that during peak load, the model's latency increases significantly, leading to a degraded user experience. The team wants to ensure that the prediction latency remains consistent even under varying traffic patterns. Which SageMaker feature should they implement to automatically adjust the inference capacity based on demand?
- ASageMaker Pipelines
- BSageMaker Model Monitor
- CSageMaker Endpoint Auto Scaling
- DSageMaker Neo
Show answer & explanationAnswer & explanation
Correct answer: C. SageMaker Endpoint Auto Scaling
SageMaker Endpoint Auto Scaling automatically adjusts the number of instances supporting a SageMaker endpoint based on predefined scaling policies. This ensures that the inference capacity matches the demand, maintaining consistent latency and performance during traffic fluctuations.
Why the other options are wrong
- A. SageMaker Pipelines is an MLOps service for creating, automating, and managing end-to-end ML workflows, not for real-time scaling of endpoints.
- B. SageMaker Model Monitor is used for detecting data drift and model quality issues, not for managing inference capacity.
- D. SageMaker Neo is used for optimizing ML models for deployment on various hardware targets, not for dynamic scaling of inference endpoints.
SageMaker Endpoint Auto Scaling
A SageMaker feature that automatically adjusts the number of inference instances for an endpoint based on traffic and performance metrics.
- Ensures consistent latency and throughput
- Reduces operational overhead by automating scaling
- Can scale based on CPU utilization, memory utilization, or custom metrics
Memory trick: Auto-scale for smooth traffic flow.