Question 686 of 1,000
hardMultiple ChoiceObjective-mapped

Accelerating Real-Time Inference with GPU in SageMaker

This MLA-C01 practice question tests your understanding of auto-scaling. Read the scenario carefully and evaluate each option against the stated constraints before committing to an answer. A key principle to apply: auto-scaling. Once you have made your selection, read the full explanation to reinforce the concept and understand why each distractor is designed to mislead on exam day.

A data science team at a financial services company is deploying a real-time fraud detection model using Amazon SageMaker. The model is a gradient boosting classifier trained on historical transaction data. The model is deployed to a SageMaker endpoint with an ML.M5.LARGE instance for real-time inference. After deployment, the team observes that the endpoint's latency spikes to over 2 seconds during peak hours (10:00-12:00 and 14:00-16:00), causing timeouts for client applications. The average latency during off-peak hours is 200 ms. The team has enabled auto-scaling with a target average CPU utilization of 70%, but the endpoint still experiences high latency during peak hours. The instance count never scales beyond 2 instances during peaks. The model size is 500 MB, and each request includes 200 features. The team needs to reduce latency to under 500 ms at the 99th percentile during peak hours without increasing costs beyond the current budget. Which course of action should the team take?

Clue words in this question

Noticing these words before you look at the options changes how you read each choice.

  • Clue: "never"

    Why it matters: Absolute qualifier. True only if the statement has zero exceptions — be cautious of options that seem obvious but break down in edge cases.

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

Increase the auto-scaling maximum instance count to 10 and set target CPU utilization to 50%.

Option B is correct because the root cause of high latency during peak hours is insufficient compute capacity. By increasing the auto-scaling maximum instance count to 10 and lowering the target CPU utilization to 50%, the endpoint will scale out more aggressively during peak traffic, distributing the inference load across more instances. This reduces per-instance CPU utilization and latency without resorting to more expensive GPU instances. The current budget likely supports up to 10 instances, so costs remain within budget. Option C is incorrect because gradient boosting inference is CPU-bound and does not benefit significantly from GPU acceleration; GPU instances are also more expensive, potentially increasing costs.

Key principle: Auto-scaling

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

  • Configure SageMaker batch transform for the real-time endpoint to process requests asynchronously.

    Why it's wrong here

    Incorrect. SageMaker batch transform is designed for offline batch processing, not for real-time inference. Using it for a real-time endpoint would not reduce latency and would break the real-time requirement.

  • Increase the auto-scaling maximum instance count to 10 and set target CPU utilization to 50%.

    Why this is correct

    Correct. Increasing the maximum instance count and lowering the CPU utilization target allows the endpoint to scale out to more instances during peak hours, distributing the workload and reducing latency. This addresses the compute bottleneck without requiring GPU instances or incurring extra costs if the budget accommodates the higher maximum.

    Clue confirmation

    The clue word "never" in the question point toward this answer.

    Related concept

    Auto-scaling

  • Switch the endpoint instance type to a GPU instance such as ml.g4dn.xlarge to accelerate inference.

    Why it's wrong here

    Incorrect. Gradient boosting models are typically CPU-bound and do not achieve significant speedups on GPU instances. GPU instances are also more expensive per hour, which could increase costs beyond the current budget.

  • Enable data compression on the endpoint to reduce payload size and network latency.

    Why it's wrong here

    Incorrect. Data compression reduces payload size and network latency, but the primary bottleneck during peak hours is compute capacity (CPU), not network. Compressing data would not sufficiently reduce inference latency to under 500 ms at the 99th percentile.

Common exam traps

Common exam trap: answer the scenario, not the keyword

The trap is that candidates might assume GPU acceleration is the standard fix for high latency, but gradient boosting models are CPU-bound. Horizontal scaling (more instances) is the appropriate and cost-effective solution.

Detailed technical explanation

How to think about this question

Gradient boosting classifiers, especially those implemented in frameworks like XGBoost or LightGBM, rely heavily on sequential tree traversal and floating-point operations, which are CPU-intensive. GPU instances leverage thousands of CUDA cores to parallelize these operations, significantly reducing inference latency for models of this size. In practice, switching to a GPU instance can yield 5-10x speedup for tree-based models, making it a cost-effective solution when CPU utilization is already saturated.

KKey Concepts to Remember

  • Auto-scaling
  • Horizontal scaling
  • Gradient boosting inference
  • Latency budget

TExam Day Tips

  • Watch for words such as best, first, most likely and least administrative effort.
  • Review why wrong options are wrong, not only why the correct option is correct.

Key takeaway

Auto-scaling

Real-world example

How this comes up in practice

A startup's cloud architect reviews their monthly bill and notices costs are higher than expected for a long-running batch job. Switching from on-demand instances to Reserved Instances — or using Spot/Preemptible VMs — can reduce compute costs by up to 72 %. Questions like this test whether you understand the tradeoffs between commitment, flexibility, and cost across cloud pricing models.

Related practice questions

Related MLA-C01 practice-question pages

Use these pages to review the topic behind this question. This is how one missed question becomes focused revision.

Practice this exam

Start a free MLA-C01 practice session

Short sessions build daily habit. Longer sessions build exam-day stamina. Try a timed session to simulate real conditions.

FAQ

Questions learners often ask

What does this MLA-C01 question test?

Auto-scaling

What is the correct answer to this question?

The correct answer is: Increase the auto-scaling maximum instance count to 10 and set target CPU utilization to 50%. — Option B is correct because the root cause of high latency during peak hours is insufficient compute capacity. By increasing the auto-scaling maximum instance count to 10 and lowering the target CPU utilization to 50%, the endpoint will scale out more aggressively during peak traffic, distributing the inference load across more instances. This reduces per-instance CPU utilization and latency without resorting to more expensive GPU instances. The current budget likely supports up to 10 instances, so costs remain within budget. Option C is incorrect because gradient boosting inference is CPU-bound and does not benefit significantly from GPU acceleration; GPU instances are also more expensive, potentially increasing costs.

What should I do if I get this MLA-C01 question wrong?

Review auto-scaling, then practise related MLA-C01 questions on the same topic to reinforce the concept.

Are there clue words in this question I should notice?

Yes — watch for: "never". Absolute qualifier. True only if the statement has zero exceptions — be cautious of options that seem obvious but break down in edge cases.

What is the key concept behind this question?

Auto-scaling

About these practice questions

Courseiva creates original exam-style practice questions with explanations and wrong-answer analysis. It does not publish real exam questions, exam dumps, or protected exam content. Learn why practice questions differ from exam dumps →

How Courseiva writes practice questions · Editorial policy

Keep practising

More MLA-C01 practice questions

Last reviewed: Jun 24, 2026

Question Discussion

Share a tip, memory trick, or ask about the reasoning behind this question. Do not post real exam questions, leaked content, braindumps, or copyrighted exam material. Comments are moderated and may be removed without notice.

Loading comments…

Sign in to join the discussion.

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.