easyMultiple Choice
MLA-C01 Practice Question: A company has a SageMaker endpoint that uses a…
A company has a SageMaker endpoint that uses a trained model to classify images. The endpoint is experiencing high latency and the team suspects it is due to the model size. Which action can the team take to reduce latency without significantly impacting accuracy?
⚠ Common exam trap
AWS often tests the misconception that converting to an open format like ONNX inherently optimizes performance, when in reality it is just a serialization format and requires a separate compilation step (e.g., Neo) to reduce latency.
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 SageMaker Neo to compile the model for the target instance
SageMaker Neo compiles trained models into an optimized binary for the target hardware, applying techniques like operator fusion, memory layout optimization, and quantization. This reduces model size and inference latency while preserving accuracy, making it the correct choice for addressing high latency caused by model size.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Switch to a compute-optimized instance type
Why it's wrong here
A compute-optimised instance adds CPU throughput but the model still occupies the same memory and executes the same operations, so size-induced latency remains. It is tempting because faster hardware often masks slow inference, and it would be correct when latency stems from CPU-bound preprocessing or insufficient vCPUs rather than model size.
- ✓
Use SageMaker Neo to compile the model for the target instance
Why this is correct
SageMaker Neo compiles the model into optimised machine code for the target instance family, cutting inference latency through graph-level and operator-level optimisations while preserving accuracy. This directly addresses the model-size-driven latency without retraining or changing the endpoint.
- ✗
Reduce the batch size of inference requests
Why it's wrong here
Batch size governs how many requests are grouped per inference call, not the model's parameter count or memory footprint, so it cannot shrink the size-driven latency. It is tempting because larger batches do raise throughput, and tuning batch size is the standard lever when latency stems from underutilised compute rather than oversized weights.
- ✗
Convert the model to ONNX format
Why it's wrong here
ONNX is an interchange format for portability across runtimes; converting alone does not compress weights or reduce parameter count, so the size-driven latency persists. It is tempting because ONNX Runtime can accelerate inference, and it would be the right choice when moving a model between frameworks or onto non-native hardware.
Go deeper
Related to this question
About these practice questions
One of 665 original MLA-C01 practice questions on Courseiva, each with a full explanation and wrong-answer analysis — not exam dumps or protected exam content. Learn why practice questions differ from exam dumps →
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
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.