MLS-C01 Modeling Practice Question

A machine learning team is using Amazon SageMaker to train a model with a large dataset stored in S3. The training job is taking too long. Which THREE of the following actions can reduce training time? (Choose three.)

⚠ Common exam trap

It's easy for candidates to confuse batch size adjustments as a primary performance lever, but the exam tests understanding that hardware upgrades (GPU power), parallelism (distributed training), and data streaming (Pipe mode) are the most direct and reliable methods to reduce training time in SageMaker.

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

✓

Use a GPU instance with more powerful GPUs.

Using a GPU instance with more powerful GPUs (Option B) reduces training time because it increases the parallel compute capacity for matrix operations, which are the core of deep learning. Amazon SageMaker allows you to select instances like p3.16xlarge with NVIDIA V100 GPUs, which offer significantly higher FLOPS compared to smaller GPU instances, directly accelerating model training.

Answer analysis

Option-by-option breakdown

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

✗
Decrease the batch size.
Why it's wrong here
Decreasing the batch size typically increases the total number of training iterations required to process the entire dataset, thereby extending the overall training time, not reducing it. This action also leads to more frequent model updates, which can add computational overhead. However, reducing batch size is often practised to mitigate out-of-memory errors on the training instance, improve model generalisation by providing noisier gradients, or accelerate convergence in certain optimisation landscapes. It would be a relevant choice if the issue were memory exhaustion or suboptimal model performance rather than excessive training duration.
✓
Use a GPU instance with more powerful GPUs.
Why this is correct
Faster GPUs reduce computation time.
✓
Use distributed training with multiple instances.
Why this is correct
Distributed training parallelizes computation across instances.
✓
Use Pipe input mode instead of File mode for the training data.
Why this is correct
Pipe mode streams data directly from S3, reducing download time.
✗
Increase the batch size.
Why it's wrong here
Increasing batch size can reduce the number of updates but may require more memory and not always reduce time.

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

Go deeper

MLS-C01 Modeling Practice Question

Why each option matters

Option-by-option breakdown

AWS S3 Storage Class Comparison

Related to this question