MLA-C01 ML Model Development Practice Question
A machine learning engineer is training a deep learning model on SageMaker using the PyTorch estimator. The training job fails with an error indicating that the GPU memory is exhausted. The engineer wants to reduce memory usage without changing the model architecture. Which SageMaker feature should the engineer use?
⚠ Common exam trap
Test-takers frequently confuse data parallelism with model parallelism; data parallelism replicates the model and does not reduce per-GPU memory usage.
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
✓
SageMaker model parallelism
SageMaker model parallelism partitions a large model across multiple GPUs, reducing the memory required on each GPU. This directly addresses GPU memory exhaustion without changing the model architecture. Data parallelism replicates the model, which does not help with memory constraints, and the other services are for monitoring or tuning, not memory reduction.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
SageMaker data parallelism
Why it's wrong here
SageMaker data parallelism distributes the training data across multiple GPUs but replicates the model on each GPU. This increases throughput but does not reduce the memory required for the model on each GPU. Since the issue is GPU memory exhaustion due to model size, data parallelism will not solve the problem and may even exacerbate it by adding communication overhead.
- ✗
SageMaker Debugger
Why it's wrong here
SageMaker Debugger is used to monitor and profile training jobs, such as detecting vanishing gradients or resource utilization. It does not reduce memory usage during training. While it can help identify memory bottlenecks, it does not provide a mechanism to alleviate GPU memory exhaustion.
- ✗
SageMaker Automatic Model Tuning
Why it's wrong here
SageMaker Automatic Model Tuning searches for optimal hyperparameters but does not address GPU memory limitations. Tuning might find a smaller batch size that reduces memory, but that is a hyperparameter change and not a direct feature to reduce memory. Moreover, it does not guarantee a solution and is not the primary purpose of the service.
- ✓
SageMaker model parallelism
Why this is correct
SageMaker model parallelism allows training of large models by partitioning the model across multiple GPUs. This reduces the memory footprint on each GPU, enabling training of models that would otherwise not fit. It is specifically designed to address memory constraints without altering the model architecture, making it the correct choice for this scenario.
Go deeper
Related to this question
About these practice questions
Courseiva writes every MLA-C01 question from scratch — 665 in total, each with an explanation and a wrong-answer breakdown. None are copied from real exams or 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.