Question 189 of 1,672
Reducing Training Time on SageMaker
A data scientist is using SageMaker to train a deep learning model. The training job runs on a single GPU instance and is taking too long. Which THREE actions can the data scientist take to reduce training time? (Choose three.)
Quick Answer
The answer is to use distributed training across multiple GPU instances, enable Pipe input mode, and choose a larger GPU instance type. Distributed training splits the workload across several GPUs, directly reducing wall-clock time through parallel computation. Pipe input mode is critical because it streams data directly from Amazon S3 to the GPU, eliminating the I/O bottleneck caused by downloading the entire dataset to an EBS volume first—this keeps the GPU busy processing rather than waiting for data, which is especially important for large datasets that exceed local memory. On the AWS Certified Machine Learning Specialty MLS-C01 exam, this question tests your understanding of SageMaker’s optimization features for deep learning; a common trap is to overlook Pipe mode in favor of File mode, or to forget that simply adding more GPUs without addressing data loading still leaves the GPU idle. Remember the mnemonic “Pipe it, split it, size it up”—Pipe mode, distributed training, and a larger instance—to quickly recall the three levers for reducing training time.
⚠ Common exam trap
Many exam-takers confuse increasing instance count (Option B) with distributed training (Option E), not realizing that simply adding instances without distributed training code and configuration does not parallelize the workload and can even increase overhead.
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
✓
Switch to Pipe input mode to reduce I/O waiting time.
Pipe input mode streams training data directly from Amazon S3 to the GPU instance, reducing I/O waiting time compared to the default File mode, which downloads the entire dataset to the EBS volume first. This minimizes the time the GPU spends idle waiting for data, especially for large datasets that cannot fit entirely in memory.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Increase the size of the EBS volume attached to the instance.
Why it's wrong here
EBS volume size does not affect compute speed.
- ✗
Increase the number of instances but keep the same total data.
Why it's wrong here
Without distributed training, adding instances doesn't help.
- ✓
Switch to Pipe input mode to reduce I/O waiting time.
Why this is correct
Streams data directly, reducing I/O bottleneck.
- ✓
Use a larger GPU instance type, such as p3.16xlarge.
Why this is correct
More GPU memory and compute power.
- ✓
Use distributed training across multiple GPU instances.
Why this is correct
Leverages parallelism to speed up training.
Quick reference
AWS S3 Storage Class Comparison
| Storage Class | Min Duration | Retrieval | Use Case |
|---|---|---|---|
| S3 Standard | None | Immediate | Frequently accessed data |
| S3 Standard-IA | 30 days | Immediate | Infrequent access, rapid retrieval |
| S3 One Zone-IA | 30 days | Immediate | Non-critical infrequent data |
| S3 Intelligent-Tiering | None | Immediate–hours | Unknown or changing access patterns |
| S3 Glacier Instant | 90 days | Milliseconds | Archive with instant retrieval |
| S3 Glacier Flexible | 90 days | Minutes–hours | Archive, flexible retrieval |
| S3 Glacier Deep Archive | 180 days | Hours | Long-term compliance archive |
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 →
Same concept, more angles
3 more ways this is tested on MLS-C01
These questions test the same concept from different angles. Work through them to make sure you can recognise it however the exam phrases it.
Variation 1. A data scientist is training a deep learning model for object detection using Amazon SageMaker. The training job is using a single GPU instance and is taking too long. Which THREE actions can reduce training time? (Choose THREE.)
medium- A.Use a CPU instance instead of GPU
- ✓ B.Enable mixed precision training with FP16
- ✓ C.Use a GPU instance with more GPUs, such as p3.16xlarge
- D.Reduce the batch size
- ✓ E.Use distributed training across multiple instances
Why B: Enabling mixed precision training with FP16 reduces memory usage and accelerates computation by using half-precision floating-point numbers where possible, which is particularly effective on NVIDIA GPUs with Tensor Cores (e.g., V100, A100). This can nearly double throughput for deep learning models without sacrificing model accuracy, as critical operations still use FP32 precision.
Variation 2. A data scientist is using SageMaker to train a deep learning model with a large dataset stored in S3. The training is taking a long time. Which action would most likely reduce training time without sacrificing accuracy?
medium- A.Increase the batch size
- ✓ B.Use SageMaker Pipe Input mode
- C.Use a smaller instance type
- D.Reduce the number of epochs
Why B: SageMaker Pipe Input mode streams training data directly from S3 into the algorithm without first downloading it to the local EBS volume. This eliminates the I/O bottleneck caused by large dataset downloads, significantly reducing training time while preserving accuracy because the model sees the same data.
Variation 3. A data scientist is using SageMaker to train a deep learning model for image classification. The training job is taking too long. Which approach can reduce training time?
medium- ✓ A.Use SageMaker's distributed data parallelism
- B.Use SageMaker Neo to compile the model
- C.Increase the number of epochs
- D.Use a smaller image size
Why A: SageMaker's distributed data parallelism splits the training data across multiple GPUs or instances, allowing each worker to process a different subset of the data simultaneously. This reduces the wall-clock time per epoch by parallelizing the computation, which directly addresses the 'taking too long' issue for deep learning image classification models.
Last reviewed: Jun 24, 2026
This MLS-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 MLS-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.