mediumMultiple Choice
MLA-C01 Practice Question: A team deploys a PyTorch model on Amazon…
A team deploys a PyTorch model on Amazon SageMaker for real-time inference. They notice that inference latency is higher than expected. They suspect the serialization format used for input data is inefficient. Which approach would MOST likely reduce latency?
⚠ Common exam trap
Many candidates confuse throughput improvements (scaling, larger instances) with latency reduction, or mistakenly think Batch Transform can substitute for real-time inference, when the question specifically targets the serialization format as the suspected bottleneck.
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
✓
Change the input serialization format to Protocol Buffers.
Protocol Buffers (protobuf) are a binary serialization format that is significantly more compact and faster to parse than text-based formats like JSON or CSV. By reducing the size of the input data and the CPU overhead of deserialization, switching to protobuf directly addresses the root cause of high inference latency on SageMaker real-time endpoints.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Use Amazon SageMaker Batch Transform instead of real-time inference.
Why it's wrong here
Batch Transform processes entire datasets asynchronously, so it cannot serve the low-latency, per-request responses the real-time endpoint requires. It is tempting because it genuinely suits large offline scoring jobs where throughput matters and individual response time does not.
- ✓
Change the input serialization format to Protocol Buffers.
Why this is correct
Protocol Buffers encode payloads as compact binary, cutting serialisation and parsing overhead compared with JSON or CSV text. Smaller request bodies also reduce network transfer time. This directly targets the inefficient input serialisation the team suspects is inflating real-time inference latency on SageMaker.
- ✗
Enable automatic scaling on the endpoint.
Why it's wrong here
Automatic scaling adds or removes endpoint instances in response to traffic; it cannot alter how a single request's payload is deserialised, so per-invocation latency stays unchanged. It is tempting because scaling genuinely helps when latency stems from queueing under concurrent load, which is not the serialisation bottleneck described here.
- ✗
Increase the instance type to a compute-optimized instance.
Why it's wrong here
Instance type changes CPU/memory capacity, not the serialization format the stem identifies as the suspected cause. Compute-optimised instances suit CPU-bound training or throughput-heavy workloads, but here the input encoding itself must change — for example to a faster format — before hardware scaling helps.
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 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.