easyMultiple Choice
Why Model Predictions Differ After Traffic Split in Vertex AI | Google PDE
A data scientist wants to test a new model version on a small percentage of traffic before full rollout. Which Vertex AI feature allows this?
⚠ Common exam trap
Candidates often confuse the conceptual practice of 'canary deployments' (Option D) with the specific Vertex AI feature 'endpoint traffic splitting' (Option B), but the exam expects the exact feature name as defined in the Google Cloud documentation.
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
✓
Endpoint traffic splitting
Vertex AI Endpoint traffic splitting allows you to route a specified percentage of inference requests to different model versions deployed on the same endpoint. This enables gradual rollout by directing a small fraction of traffic (e.g., 5%) to the new model while the rest goes to the current version, without needing separate endpoints or manual routing logic.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
A/B testing
Why it's wrong here
A/B testing splits traffic between model versions to measure performance, but it does not itself limit rollout to a small percentage before full deployment — that is traffic splitting on a Vertex AI endpoint. It is tempting because A/B testing is genuinely used for controlled model comparison experiments.
- ✓
Endpoint traffic splitting
Why this is correct
Endpoint traffic splitting lets you assign percentage weights to multiple deployed model versions behind one endpoint, routing a small share of requests to the new version. This directly satisfies the gradual-rollout constraint before promoting it to full traffic.
- ✗
Model monitoring
Why it's wrong here
Model monitoring detects skew, drift and prediction quality on deployed models; it does not route a percentage of requests to a new version. Traffic splitting on an endpoint performs that gradual exposure. Monitoring is tempting because it also supports safe rollout, but it observes after deployment rather than controlling request distribution.
- ✗
Model versioning with canary deployments
Why it's wrong here
Vertex AI model versioning tracks artefacts and metadata; it does not split live traffic between versions. Canary routing is delivered through traffic-split configuration on an endpoint, not the version registry itself. Versioning is tempting because it manages rollback and lineage, which is a different concern from gradual exposure.
Go deeper
Related to this question
About these practice questions
One of 747 original PDE 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 →
Same concept, more angles
1 more way this is tested on PDE
These questions test the same concept from different angles. Work through them to make sure you can recognise it however the exam phrases it.
Variation 1. A data scientist has iterated on a model and produced a new version. The organization requires the ability to roll back to the previous version quickly if the new version performs poorly in production. Which approach should be used?
easy- A.Store each model version in a separate Cloud Storage bucket.
- B.Keep the previous model in a container image and redeploy via Cloud Run.
- C.Use Cloud Source Repositories to tag model versions.
- ✓ D.Upload both versions to Vertex AI Model Registry and use endpoint traffic splitting to route 100% to the safe version if needed.
Why D: Vertex AI Model Registry allows you to deploy multiple model versions and use endpoint traffic splitting to gradually shift traffic or instantly route 100% to a specific version. This enables immediate rollback by setting the traffic split to 100% for the previous model version without redeploying or changing infrastructure.
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
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.