- A
Deploy multiple versions of the model to the same endpoint using traffic splitting and set the primary version to 100% traffic.
Why wrong: Traffic splitting is used for canary deployments, not for versioning and rollback management.
- B
Use the same model name for all deployments and overwrite the existing model.
Why wrong: Overwriting models loses previous versions, making rollback impossible.
- C
Store the model in a Cloud Storage bucket with a fixed name and rely on Cloud Build for rollback.
Why wrong: A fixed name does not provide versioning; Cloud Build does not automatically manage model version rollbacks.
- D
Create a new model resource in Vertex AI for each version and deploy the specific version to an endpoint.
This allows version tracking, easy rollback by redeploying a previous version, and maintains a clean deployment history.
Quick Answer
The correct practice is to create a new model resource in Vertex AI for each version and deploy the specific version to an endpoint. This approach ensures traceability and rollback capability because each model resource is independently tracked with a unique ID, allowing you to deploy or undeploy a specific version without affecting others. On the Google Professional Data Engineer exam, this tests your understanding of Vertex AI’s deployment architecture, where the model resource acts as a versioned container separate from the endpoint. A common trap is thinking you can simply overwrite the model file in Cloud Storage or update an existing deployment, but that breaks version history and makes rollback impossible. Remember the key distinction: in Vertex AI, the model resource is the version, not the file in the bucket. Memory tip: “One model resource per version, one version per endpoint deployment.”
PDE Operationalizing machine learning models Practice Question
This PDE practice question tests your understanding of operationalizing machine learning 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 deploys a new machine learning model for real-time predictions using Vertex AI. The model is stored in a Cloud Storage bucket and deployed to an endpoint. To ensure traceability and rollback capability, which practice should be followed?
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
Create a new model resource in Vertex AI for each version and deploy the specific version to an endpoint.
Option D is correct because creating a new model resource in Vertex AI for each version ensures that each model iteration is independently tracked, versioned, and can be deployed to an endpoint with full rollback capability. This practice aligns with Vertex AI's model versioning and endpoint deployment model, where each model resource has a unique ID and can be deployed or undeployed without affecting other versions, enabling precise traceability and rollback.
Key principle: Answer the scenario, not the keyword: identify the specific constraint before choosing the most familiar-sounding option.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Deploy multiple versions of the model to the same endpoint using traffic splitting and set the primary version to 100% traffic.
Why it's wrong here
Traffic splitting is used for canary deployments, not for versioning and rollback management.
- ✗
Use the same model name for all deployments and overwrite the existing model.
Why it's wrong here
Overwriting models loses previous versions, making rollback impossible.
- ✗
Store the model in a Cloud Storage bucket with a fixed name and rely on Cloud Build for rollback.
Why it's wrong here
A fixed name does not provide versioning; Cloud Build does not automatically manage model version rollbacks.
- ✓
Create a new model resource in Vertex AI for each version and deploy the specific version to an endpoint.
Why this is correct
This allows version tracking, easy rollback by redeploying a previous version, and maintains a clean deployment history.
Related concept
Read the scenario before looking for a memorised answer.
Common exam traps
Common exam trap: answer the scenario, not the keyword
Google Cloud often tests the misconception that traffic splitting alone (Option A) provides sufficient versioning and rollback, but the trap is that traffic splitting still operates within a single model resource, which does not preserve independent version history or allow clean rollback to a prior model resource without manual intervention.
Detailed technical explanation
How to think about this question
In Vertex AI, each model resource is an independent entity with its own ID, version, and metadata (e.g., artifact URI, training details). When deploying to an endpoint, you can deploy multiple model resources and use traffic splitting (e.g., 90/10) for canary testing, but for full rollback, you simply undeploy the problematic model and deploy a previous model resource. Under the hood, Vertex AI uses the model resource's artifact URI to load the model from Cloud Storage, so keeping each version as a separate resource ensures the artifact is immutable and traceable via the model's metadata.
KKey Concepts to Remember
- Read the scenario before looking for a memorised answer.
- Find the constraint that changes the correct option.
- Eliminate answers that are true in general but not in this case.
TExam Day Tips
- Watch for words such as best, first, most likely and least administrative effort.
- Review why wrong options are wrong, not only why the correct option is correct.
Key takeaway
Answer the scenario, not the keyword: identify the specific constraint before choosing the most familiar-sounding option.
Real-world example
How this comes up in practice
A media company stores terabytes of video archives that are accessed once a year for audit purposes. Moving these objects to a cold storage tier (Azure Archive, S3 Glacier, or Google Nearline) costs a fraction of hot storage. Questions like this test whether you understand storage tiers, access frequency tradeoffs, and retrieval latency requirements.
What to study next
Got this wrong? Here's your next step.
Identify which exam domain this question belongs to, review the core concept, then practise similar questions from the same domain.
- →
Operationalizing machine learning models — study guide chapter
Learn the concepts, then practise the questions
- →
Operationalizing machine learning models practice questions
Targeted practice on this topic area only
- →
All PDE questions
499 questions across all exam domains
- →
Google Professional Data Engineer study guide
Full concept coverage aligned to exam objectives
- →
PDE practice test guide
How to use practice tests most effectively before exam day
Related practice questions
Related PDE practice-question pages
Use these pages to review the topic behind this question. This is how one missed question becomes focused revision.
Designing data processing systems practice questions
Practise PDE questions linked to Designing data processing systems.
Building and operationalizing data processing systems practice questions
Practise PDE questions linked to Building and operationalizing data processing systems.
Operationalizing machine learning models practice questions
Practise PDE questions linked to Operationalizing machine learning models.
Ensuring solution quality practice questions
Practise PDE questions linked to Ensuring solution quality.
PDE fundamentals practice questions
Practise PDE questions linked to PDE fundamentals.
PDE scenario practice questions
Practise PDE questions linked to PDE scenario.
PDE troubleshooting practice questions
Practise PDE questions linked to PDE troubleshooting.
Practice this exam
Start a free PDE practice session
Short sessions build daily habit. Longer sessions build exam-day stamina. Try a timed session to simulate real conditions.
FAQ
Questions learners often ask
What does this PDE question test?
Operationalizing machine learning models — This question tests Operationalizing machine learning models — Read the scenario before looking for a memorised answer..
What is the correct answer to this question?
The correct answer is: Create a new model resource in Vertex AI for each version and deploy the specific version to an endpoint. — Option D is correct because creating a new model resource in Vertex AI for each version ensures that each model iteration is independently tracked, versioned, and can be deployed to an endpoint with full rollback capability. This practice aligns with Vertex AI's model versioning and endpoint deployment model, where each model resource has a unique ID and can be deployed or undeployed without affecting other versions, enabling precise traceability and rollback.
What should I do if I get this PDE question wrong?
Identify which exam domain this question belongs to, review the core concept, then practise similar questions from the same domain.
What is the key concept behind this question?
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 →
Last reviewed: Jun 30, 2026
This PDE 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 PDE 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.