MLA-C01 Practice Question: ML Solution Monitoring, Maintenance, and Security
A company uses SageMaker endpoints for real-time inference. They want to automatically scale the number of instances based on the number of outstanding requests. Which auto-scaling policy type should they choose?
⚠ Common exam trap
MLA-C01 often tests the confusion between metric-driven and schedule-driven scaling — candidates who see 'outstanding requests' and pick scheduled or step scaling miss that target tracking is the AWS-recommended default for utilization metrics.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Target tracking scaling
Target tracking scaling is the correct choice because it lets you specify a target value for a metric — such as a custom metric for outstanding requests per instance — and SageMaker automatically adjusts instance count to keep that metric at the target. This is the recommended policy for metrics that correlate directly with capacity needs, like request backlog or invocations per instance. It handles both scale-out and scale-in automatically without manual threshold tuning.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Scheduled scaling
Why it's wrong here
Scheduled scaling changes instance counts at predefined times, so it cannot react to the live number of outstanding requests. It suits predictable traffic patterns, such as known daily peaks. Target tracking scaling, driven by the backlog metric, satisfies this dynamic requirement.
- ✗
Step scaling
Why it's wrong here
Step scaling adjusts capacity in discrete tiers when a CloudWatch alarm breaches, reacting to aggregated metrics rather than queue depth directly. It suits predictable load bands. Scaling on outstanding requests requires a target tracking policy on the custom metric.
- ✓
Target tracking scaling
Why this is correct
Target tracking scaling adjusts instance count to hold a chosen metric, such as SageMakerVariantInvocationsPerInstance, at a target value, which directly reflects outstanding request load. Step and scheduled policies react to fixed thresholds or times rather than demand.
- ✗
Simple scaling
Why it's wrong here
Simple scaling adjusts capacity by a fixed increment after a CloudWatch alarm breaches, so it cannot track the continuously varying number of outstanding requests. It suits steady, predictable step changes. Target tracking scaling, which maintains a metric at a target value, matches this backlog-based requirement.
Go deeper
Related to this question
About these practice questions
This MLA-C01 question is part of Courseiva's 665-question bank — original exam-style content with full explanations and wrong-answer analysis, never real exam questions or exam dumps. Learn why practice questions differ from exam dumps →
JA
Written and reviewed by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
Last reviewed September 2026 · checked against the official Amazon Web Services exam blueprint
This MLA-C01 practice question is part of Courseiva's free Amazon Web Services certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the MLA-C01 exam.