mediumMultiple ChoiceObjective-mapped
Why Model Predictions Differ After Traffic Split in Vertex AI | Google PDE
Exhibit
Refer to the exhibit.
```
$ gcloud ai endpoints describe my-endpoint
...
trafficSplit:
model1: 70
model2: 30
...
$ gcloud ai models describe model1
...
containerSpec:
imageUri: us-central1-docker.pkg.dev/my-project/my-repo/model1:v1
env:
- name: MODEL_NAME
value: fraud_detection_v1
...
$ gcloud ai models describe model2
...
containerSpec:
imageUri: us-central1-docker.pkg.dev/my-project/my-repo/model2:v1
env:
- name: MODEL_NAME
value: fraud_detection_v2
...
```A data scientist deploys a new version of a fraud detection model (model2) alongside the existing model (model1) on the same Vertex AI endpoint with a 70/30 traffic split. After 24 hours, the team notices that model2's predictions are significantly different from model1's, and the fraud detection rate has increased. What is the most likely explanation for the change in predictions?
Quick Answer
The answer is that model2 uses a different model artifact (fraud_detection_v2) that inherently produces different predictions. This is correct because in Vertex AI, a traffic split routes the same live input data to both model versions, but each model artifact processes that data independently through its own trained logic. If model2’s predictions shift significantly and the fraud detection rate rises, the root cause is the artifact itself—not data drift, routing errors, or environment differences. On the Google Professional Data Engineer exam, this question tests your understanding that a traffic split isolates model behavior, so prediction differences directly reflect artifact changes, not infrastructure issues. A common trap is to blame data skew or endpoint configuration, but the key is that both models receive identical requests. Memory tip: “Same data, different artifact = different output.”
⚠ Common exam trap
Google Cloud often tests the misconception that a traffic split or routing issue can cause prediction differences, when in fact the split only controls which model receives the request, not the content of the request or the model's internal logic.
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
✓
Model2 uses a different model artifact (fraud_detection_v2) that produces different predictions.
The most straightforward explanation for a significant change in predictions and an increased fraud detection rate is that model2 uses a different model artifact (fraud_detection_v2) that was designed to produce different outputs. In Vertex AI, deploying a new model version with a traffic split means both models receive the same input data, but each model artifact independently processes it. If model2's predictions differ substantially, it indicates the model artifact itself has been updated or replaced, not that there is a data or routing issue.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Model2 was trained on data that leaked future information, causing unrealistic results.
Why it's wrong here
No evidence of data leakage; the difference is due to model version.
- ✗
Model2 is receiving corrupted input data due to a bug in the traffic routing.
Why it's wrong here
The traffic split is functioning as configured; no routing bug.
- ✗
The traffic split is misconfigured and sending all traffic to model2.
Why it's wrong here
The exhibit shows a 70/30 split, so model1 still receives majority traffic.
- ✓
Model2 uses a different model artifact (fraud_detection_v2) that produces different predictions.
Why this is correct
The environment variable MODEL_NAME points to different model versions, causing output differences.
Go deeper
Related to this question
About these practice questions
One of 890 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
2 more ways 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.
Variation 2. 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?
easy- A.A/B testing
- ✓ B.Endpoint traffic splitting
- C.Model monitoring
- D.Model versioning with canary deployments
Why B: 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.
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.