- A
Split the dataset into smaller files
Why wrong: Splitting files can help with I/O but is not a requirement for distributed training.
- B
Use SageMaker's distributed data parallelism library
Why wrong: While SageMaker offers a library, the more general step is to modify the script; this option is too specific and redundant with A.
- C
Modify the training script to use Horovod or PyTorch DistributedDataParallel
These frameworks enable multi-GPU communication and are necessary for distributed training.
- D
Enable automatic mixed precision
Why wrong: Mixed precision speeds up training on a single GPU but does not enable distributed training across instances.
- E
Increase the number of worker instances in the training job
Distributed training requires multiple instances to parallelize the workload.
MLA-C01 Practice Question: A machine learning engineer is training a neural…
This MLA-C01 practice question tests your understanding of mla-c01 exam topics. Read the scenario carefully and evaluate each option against the stated constraints before committing to an answer. After answering, compare your reasoning against the explanation and wrong-answer breakdown below. 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 machine learning engineer is training a neural network using Amazon SageMaker. The training job uses a single GPU instance. To improve training speed using distributed training, which two steps should they take? (Select TWO.)
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
Modify the training script to use Horovod or PyTorch DistributedDataParallel
Option C is correct because distributed training on a single GPU instance requires a framework-level approach like Horovod or PyTorch DistributedDataParallel (DDP) to coordinate gradient computation across multiple GPUs. SageMaker's distributed data parallelism library (Option B) is designed for multi-instance setups, not single-instance multi-GPU scenarios. Modifying the training script to use Horovod or DDP enables efficient allreduce-based gradient synchronization, which is essential for scaling training across GPUs within a single instance.
Key principle: Answer the scenario, not the keyword: identify the specific constraint before choosing the most familiar-sounding option.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Split the dataset into smaller files
Why it's wrong here
Splitting files can help with I/O but is not a requirement for distributed training.
- ✗
Use SageMaker's distributed data parallelism library
Why it's wrong here
While SageMaker offers a library, the more general step is to modify the script; this option is too specific and redundant with A.
- ✓
Modify the training script to use Horovod or PyTorch DistributedDataParallel
Why this is correct
These frameworks enable multi-GPU communication and are necessary for distributed training.
Related concept
Read the scenario before looking for a memorised answer.
- ✗
Enable automatic mixed precision
Why it's wrong here
Mixed precision speeds up training on a single GPU but does not enable distributed training across instances.
- ✓
Increase the number of worker instances in the training job
Why this is correct
Distributed training requires multiple instances to parallelize the workload.
Related concept
Read the scenario before looking for a memorised answer.
Common exam traps
Common exam trap: answer the scenario, not the keyword
The key pitfall is assuming that SageMaker's built-in distributed data parallelism library (Option B) is the correct choice for a single multi-GPU instance, when in reality it is designed for multi-instance training. Framework-level tools like Horovod or PyTorch DDP (Option C) are needed for single-instance multi-GPU parallelism. Additionally, candidates may think that splitting data (Option A) or enabling mixed precision (Option D) constitutes distributed training, but these are separate optimizations.
Detailed technical explanation
How to think about this question
Horovod uses a ring-allreduce algorithm to efficiently synchronize gradients across GPUs, reducing communication overhead compared to parameter server approaches. PyTorch DDP leverages the NCCL backend for optimized GPU-to-GPU communication, automatically handling gradient bucketing and overlapping computation with communication. In practice, for a single-instance multi-GPU setup, DDP or Horovod can achieve near-linear speedup by distributing the batch across GPUs and synchronizing gradients after each backward pass.
KKey Concepts to Remember
- Read the scenario before looking for a memorised answer.
- Find the constraint that changes the correct option.
- Eliminate answers that are true in general but not in this case.
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
Answer the scenario, not the keyword: identify the specific constraint before choosing the most familiar-sounding option.
Real-world example
How this comes up in practice
A healthcare organisation deploys an application with a public-facing web tier and a private database tier. The database subnet has no public IP and only accepts connections from the web tier's security group. Questions like this test whether you can design cloud network isolation using VNets/VPCs, subnets, and security group rules.
What to study next
Got this wrong? Here's your next step.
Identify which exam domain this question belongs to, review the core concept, then practise similar questions from the same domain.
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.
ML Model Development practice questions
Practise MLA-C01 questions linked to ML Model Development.
Data Preparation for Machine Learning practice questions
Practise MLA-C01 questions linked to Data Preparation for Machine Learning.
Deployment and Orchestration of ML Workflows practice questions
Practise MLA-C01 questions linked to Deployment and Orchestration of ML Workflows.
ML Solution Monitoring, Maintenance, and Security practice questions
Practise MLA-C01 questions linked to ML Solution Monitoring, Maintenance, and Security.
ML Solution Monitoring, Maintenance and Security practice questions
Practise MLA-C01 questions linked to ML Solution Monitoring, Maintenance and Security.
MLA-C01 fundamentals practice questions
Practise MLA-C01 questions linked to MLA-C01 fundamentals.
MLA-C01 scenario practice questions
Practise MLA-C01 questions linked to MLA-C01 scenario.
MLA-C01 troubleshooting practice questions
Practise MLA-C01 questions linked to MLA-C01 troubleshooting.
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?
Read the scenario before looking for a memorised answer.
What is the correct answer to this question?
The correct answer is: Modify the training script to use Horovod or PyTorch DistributedDataParallel — Option C is correct because distributed training on a single GPU instance requires a framework-level approach like Horovod or PyTorch DistributedDataParallel (DDP) to coordinate gradient computation across multiple GPUs. SageMaker's distributed data parallelism library (Option B) is designed for multi-instance setups, not single-instance multi-GPU scenarios. Modifying the training script to use Horovod or DDP enables efficient allreduce-based gradient synchronization, which is essential for scaling training across GPUs within a single instance.
What should I do if I get this MLA-C01 question wrong?
Identify which exam domain this question belongs to, review the core concept, then practise similar questions from the same domain.
What is the key concept behind this question?
Read the scenario before looking for a memorised answer.
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 →
Keep practising
More MLA-C01 practice questions
- A team is using SageMaker Pipelines to train a model. The pipeline has multiple steps: data processing, training, evalua…
- A machine learning team deploys a custom container image for an Amazon SageMaker training job. The container needs to ac…
- A machine learning engineer sees the above error in Amazon CloudWatch Logs for a SageMaker endpoint. What is the most li…
- A data scientist has trained a model that achieves 95% accuracy on the training set but only 70% on the test set. Which…
- Refer to the exhibit. A data scientist reviews the output of a SageMaker training job. The model has 95% training accura…
- A team is using Amazon SageMaker to train a neural network. They want to minimize training time while effectively explor…
Last reviewed: Jul 4, 2026
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.
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.
Sign in to join the discussion.