Quick Answer
The answer is to load the vocabulary file once at startup and reuse it across all prediction requests. This is correct because loading large assets per request in Vertex AI custom containers causes memory to be consumed repeatedly, quickly exhausting the 4GB container limit and triggering out-of-memory errors that manifest as 502 gateway errors after a few hours of traffic. On the Google Professional Machine Learning Engineer exam, this scenario tests your understanding of container lifecycle management and efficient resource allocation for inference serving, with the common trap being to mistakenly debug network or scaling issues instead of recognizing the repeated memory allocation pattern. A reliable memory tip is to think of your container as a single-occupancy apartment: you unpack your heavy luggage (the 3GB vocabulary) once when you move in, not every time you open the door for a guest.
This PMLE practice question tests your understanding of serving and scaling models. 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 company serves a PyTorch model using a custom container on Vertex AI Prediction. They notice that after a few hours, the endpoint returns 502 errors. The logs show 'Out of memory' errors. The container has a memory limit of 4GB, and the model loads a 3GB vocabulary file. What is the most likely cause and best fix?
Clue words in this question
Noticing these words before you look at the options changes how you read each choice.
Clue: "best"
Why it matters: Signals that multiple options may be partially correct. Choose the option that most directly solves the exact problem described, not the one that sounds most complete.
Clue: "most likely"
Why it matters: Probability qualifier — the question wants the most probable cause or outcome, not a guaranteed one. Eliminate low-probability options.
Increase the container memory to 8GB.
Why wrong: Band-aid; memory leak may still occur.
Load the vocabulary file once at startup and reuse it.
Prevents repeated loading, solving OOM.
Increase the number of replicas to distribute load.
Why wrong: Does not fix per-instance memory issue.
Switch to Vertex AI Batch Prediction.
Why wrong: Not relevant to online serving memory issue.
Answer choices
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
Load the vocabulary file once at startup and reuse it.
The 502 errors and 'Out of memory' errors indicate that the container is running out of memory during inference. Since the model loads a 3GB vocabulary file, and the container has only 4GB of memory, loading this file repeatedly for each prediction request (e.g., inside the prediction handler) would quickly exhaust memory. The correct fix is to load the vocabulary file once at container startup and reuse it across all requests, which is a standard best practice for serving models with large static assets.
Key principle: Answer the scenario, not the keyword: identify the specific constraint before choosing the most familiar-sounding option.
Answer analysis
For each option: why learners choose it and why it is or isn't the right answer here.
Increase the container memory to 8GB.
Why it's wrong here
Band-aid; memory leak may still occur.
Load the vocabulary file once at startup and reuse it.
Why this is correct
Prevents repeated loading, solving OOM.
Clue confirmation
The clue words "best", "most likely" in the question point toward this answer.
Related concept
Read the scenario before looking for a memorised answer.
Increase the number of replicas to distribute load.
Why it's wrong here
Does not fix per-instance memory issue.
Switch to Vertex AI Batch Prediction.
Why it's wrong here
Not relevant to online serving memory issue.
Common exam traps
Google Cloud often tests the misconception that OOM errors are always solved by increasing memory, but the trap here is that the real issue is inefficient resource reuse—loading a large file per request—rather than insufficient total memory.
Detailed technical explanation
Under the hood, PyTorch model serving containers typically use a preload pattern where large artifacts (e.g., tokenizers, embedding tables) are loaded in the global scope or via a singleton class, ensuring they are initialized once per container lifecycle. In Vertex AI Prediction, the container's memory limit is enforced by cgroups, and if the model handler loads the vocabulary file on every request (e.g., inside the predict() method), memory is allocated and not freed promptly, leading to fragmentation and OOM kills. A real-world scenario is serving a BERT model with a 2GB vocabulary; failing to cache it causes rapid memory exhaustion under load.
Key takeaway
Answer the scenario, not the keyword: identify the specific constraint before choosing the most familiar-sounding option.
Real-world example
A cloud solutions architect for a retail company is evaluating services for a new workload. The correct answer here reflects best practice for the specific scenario described — not a general cloud recommendation. Answer the scenario, not the keyword: identify the specific constraint before choosing the most familiar-sounding option. Cloud exam questions reward reading the constraint carefully: the same technology can be right or wrong depending on the use case.
What to study next
Identify which exam domain this question belongs to, review the core concept, then practise similar questions from the same domain.
Serving and scaling models — study guide chapter
Learn the concepts, then practise the questions
Serving and scaling models practice questions
Targeted practice on this topic area only
All PMLE questions
506 questions across all exam domains
Google Professional Machine Learning Engineer study guide
Full concept coverage aligned to exam objectives
PMLE practice test guide
How to use practice tests most effectively before exam day
Related practice questions
Use these pages to review the topic behind this question. This is how one missed question becomes focused revision.
Practise PMLE questions linked to Scaling prototypes into ML models.
Practise PMLE questions linked to Automating and orchestrating ML pipelines.
Practise PMLE questions linked to Collaborating within and across teams to manage data and models.
Practise PMLE questions linked to Architecting low-code ML solutions.
Practise PMLE questions linked to Collaborating to manage data and models.
Practise PMLE questions linked to Serving and scaling models.
Practise PMLE questions linked to Monitoring ML solutions.
Practise PMLE questions linked to Solving business challenges with ML.
Practise PMLE questions linked to PMLE fundamentals.
Practise PMLE questions linked to PMLE scenario.
Practise PMLE questions linked to PMLE troubleshooting.
Practice this exam
Short sessions build daily habit. Longer sessions build exam-day stamina. Try a timed session to simulate real conditions.
FAQ
Serving and scaling models — This question tests Serving and scaling models — Read the scenario before looking for a memorised answer..
The correct answer is: Load the vocabulary file once at startup and reuse it. — The 502 errors and 'Out of memory' errors indicate that the container is running out of memory during inference. Since the model loads a 3GB vocabulary file, and the container has only 4GB of memory, loading this file repeatedly for each prediction request (e.g., inside the prediction handler) would quickly exhaust memory. The correct fix is to load the vocabulary file once at container startup and reuse it across all requests, which is a standard best practice for serving models with large static assets.
Identify which exam domain this question belongs to, review the core concept, then practise similar questions from the same domain.
Yes — watch for: "best", "most likely". Signals that multiple options may be partially correct. Choose the option that most directly solves the exact problem described, not the one that sounds most complete.
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
Last reviewed: Jun 30, 2026
This PMLE practice question is part of Courseiva's free Google Cloud 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 PMLE 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.