PDE Practice Test 33 — 15 Questions

Question 1

A company uses Cloud Dataflow to process financial transactions from Pub/Sub to BigQuery. The pipeline must ensure exactly-once semantics. Recently, they noticed duplicate rows in BigQuery. The source publishes with at-least-once. The Dataflow pipeline uses idempotent writes. What is the most likely cause? Options: A. The pipeline uses GlobalWindows. B. The pipeline has autoscaling enabled. C. The pipeline uses file loads as a sink. D. The pipeline's watermark is misconfigured.

Accepted Answer

The pipeline's watermark is misconfigured. The most likely cause is a misconfigured watermark. In Dataflow, the watermark tracks event time progress and determines when to trigger window results. If the watermark is misconfigured (e.g., too aggressive or based on incorrect timestamps), late-arriving data may be processed in multiple windows, leading to duplicate rows even with idempotent writes. Since the source uses at-least-once delivery, late data can be re-published, and a faulty watermark can cause it to be written again.

Answer

The pipeline uses file loads as a sink

Answer

The pipeline uses GlobalWindows

Answer

The pipeline has autoscaling enabled

Question 2

A retail company is using a machine learning model for inventory forecasting. They observe that the model's predictions become less accurate over time, especially during holiday seasons. Which monitoring metric should they prioritize?

Accepted Answer

Prediction drift (feature drift). Prediction drift (feature drift) is the correct metric because it directly measures changes in the input data distribution over time, which is the root cause of degrading model accuracy during holiday seasons. When customer behavior shifts (e.g., buying patterns during holidays), the features the model relies on drift, causing predictions to become less accurate. Monitoring prediction drift allows the team to detect when retraining or updating the model is necessary.

Answer

Model latency

Answer

Prediction counts

Answer

Resource utilization

Question 3

A company has a trained model stored in Vertex AI Model Registry. They want to automate retraining when new training data arrives in Cloud Storage. Which approach is most efficient?

Accepted Answer

Use Cloud Functions triggered by Cloud Storage events to start a Vertex AI Training job. Cloud Functions can be directly triggered by Cloud Storage events (e.g., object finalize) to invoke the Vertex AI Training service via the AI Platform API. This creates an event-driven, serverless pipeline that retrains the model immediately when new data arrives, without polling or manual intervention, making it the most efficient and cost-effective approach.

Answer

Use Dataflow to continuously update the model

Answer

Use Cloud Scheduler to trigger a Cloud Build retraining step

Answer

Schedule a weekly Cloud Composer DAG to check for new data and retrain

Question 4

You have two versions of a classification model (v1 and v2) deployed on a Vertex AI Endpoint. You want to gradually roll out v2 to 10% of traffic, monitor performance, and if metrics are better, increase traffic to 100%. You have set up model monitoring for skew and drift. Which configuration should you use?

Accepted Answer

Use the Vertex AI Endpoint 'traffic_split' parameter to assign 10% of traffic to v2 and 90% to v1.. The Vertex AI Endpoint 'traffic_split' parameter allows you to direct a percentage of inference requests to different model versions deployed on the same endpoint. Setting 10% to v2 and 90% to v1 enables a gradual rollout while monitoring skew and drift, and you can adjust the split as needed. This is the native, supported method for canary deployments in Vertex AI, avoiding the complexity and latency of external load balancers.

Answer

Deploy v2 to a separate endpoint and use a load balancer to route 10% of traffic.

Answer

Create a new deployment with v2 on the same endpoint and set the 'min_replica_count' to 1 for both versions.

Answer

Enable Vertex AI Model Monitoring on the endpoint and set up alerting for performance drop.

Question 5

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?

Accepted Answer

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

A/B testing

Answer

Model monitoring

Answer

Model versioning with canary deployments

Question 6

Refer to the exhibit. This log entry was generated by Vertex AI Model Monitoring for a production model. What should the data engineer do to address this issue?

Accepted Answer

Retrain the model with more recent data. Option B is correct because Vertex AI Model Monitoring detected a drift in the 'age' feature, indicating that the production data distribution has shifted from the training data. Retraining the model with more recent data aligns the model with the current data distribution, mitigating the drift and maintaining prediction accuracy. This is the standard remediation for model drift in production ML systems.

Answer

Increase the drift threshold to 0.9 to suppress alerts

Answer

Deploy a new model version trained on the original dataset

Answer

Disable monitoring for the 'age' feature

Question 7

You need to automate retraining of a model when new training data becomes available every week. The training pipeline runs on Vertex AI Pipelines and is triggered by Cloud Composer. After retraining, you want to evaluate the new model against a golden dataset. If the model's accuracy improves by at least 1%, it should be automatically deployed to the staging endpoint. What is the best way to implement the decision logic?

Accepted Answer

Add a conditional step in the Vertex AI Pipeline to evaluate the model and deploy if the accuracy improvement threshold is met.. Option B is correct because Vertex AI Pipelines supports conditional execution natively via the `Condition` component, allowing you to evaluate the new model's accuracy against the golden dataset within the same pipeline and deploy only if the improvement threshold (≥1%) is met. This approach keeps the entire retraining, evaluation, and deployment workflow automated, auditable, and tightly coupled within a single orchestrated pipeline, avoiding external triggers or manual steps.

Answer

Use Cloud Functions to compare metrics and call the endpoint if conditions are met.

Answer

After training, run a batch prediction job on the golden dataset and compare metrics manually.

Answer

Use Vertex AI Experiments to log metrics and set up an alert to manually deploy.

Question 8

A team is deploying a complex model with multiple preprocessing steps. They want to ensure consistent preprocessing during training and serving. Which three approaches can achieve this? (Select 3)

Accepted Answer

Store preprocessing logic in a shared Python module. Option A is correct because storing preprocessing logic in a shared Python module ensures that the same code is used during both training and serving, eliminating drift between environments. This approach leverages version control and dependency management to guarantee consistency, which is critical for reproducibility in production ML pipelines.

Answer

Use a separate preprocessing service called from the model

Answer

Use two separate pipelines for training and serving

Question 9

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?

Accepted Answer

Upload both versions to Vertex AI Model Registry and use endpoint traffic splitting to route 100% to the safe version if needed.. 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.

Answer

Store each model version in a separate Cloud Storage bucket.

Answer

Keep the previous model in a container image and redeploy via Cloud Run.

Answer

Use Cloud Source Repositories to tag model versions.

Question 10

A team is deploying a TensorFlow model for online predictions on AI Platform Prediction. They want to monitor for data drift and model performance degradation. Which TWO Google Cloud services should they use?

Accepted Answer

AI Platform Continuous Evaluation. AI Platform Continuous Evaluation (option B) is correct because it is a managed service specifically designed to detect data drift and model performance degradation in deployed models. It automatically compares incoming prediction data against the training data distribution and monitors metrics like accuracy over time, triggering alerts when significant drift is detected. Cloud Monitoring (option C) is correct because it provides the underlying metrics and alerting infrastructure that can track model performance indicators (e.g., prediction latency, error rates) and integrate with Continuous Evaluation for comprehensive observability.

Answer

Cloud Composer

Answer

AI Platform Pipelines

Answer

Cloud Logging

Question 11

A Cloud Build pipeline is set up to train a model on Vertex AI. The build fails with the error: 'ERROR: (gcloud.ai-platform.jobs.submit.training) NOT_FOUND: The parent project does not exist.' The project ID and the service account are correctly configured. What is the most likely cause?

Accepted Answer

The Cloud Build service account does not have the aiplatform.jobs.create permission on the project.. The error 'NOT_FOUND: The parent project does not exist' indicates that the Cloud Build service account lacks the necessary IAM permission to submit a training job to Vertex AI. Even though the project ID and service account are correctly configured, the Cloud Build service account must have the 'aiplatform.jobs.create' permission (or the 'Vertex AI User' role) on the project. Without this, the API call fails because the service account is not authorized to access the project resource.

Answer

The region specified for the training job does not exist.

Answer

The training job requires a GPU, which is not available in the specified region.

Answer

The training package is not uploaded to Cloud Storage before the pipeline runs.

Question 12

A company runs a real-time fraud detection model using Cloud Dataflow for streaming inference. The model is updated every hour with new training data. The team wants to minimize downtime and ensure that both old and new model versions are available during the update. Which deployment strategy should they use?

Accepted Answer

Canary deployment: deploy the new model alongside the old one, gradually increase traffic to the new model while monitoring.. Canary deployment is the correct strategy because it allows the new model to be deployed alongside the old one, with traffic gradually shifted to the new version while monitoring for errors or performance degradation. This minimizes downtime and ensures both versions are available during the update, which is critical for a real-time fraud detection system where continuous availability and risk mitigation are paramount.

Answer

A/B testing: route a small percentage of traffic to the new model and compare performance.

Answer

Rolling deployment: gradually replace instances of the old model with the new model.

Answer

Blue/green deployment: deploy the new model to a separate endpoint, then switch all traffic at once.

Question 13

Refer to the exhibit. A developer sees this log entry when trying to get a prediction. What is the most likely cause?

Accepted Answer

The model version is not deployed. The log entry indicates that the model version specified in the request is not currently deployed to the serving infrastructure. In Google Cloud's Vertex AI, a model version must be explicitly deployed to an endpoint before it can serve predictions; attempting to predict against a non-deployed version returns an error. This is the most likely cause because the error message directly references the model version's deployment status.

Answer

The model ID is incorrect

Answer

The endpoint does not exist

Answer

The project ID is wrong

Question 14

Refer to the exhibit. A data scientist notices that the evaluation component rarely passes the threshold, causing the pipeline to fail often. What should they do to improve efficiency?

Accepted Answer

Add a conditional component that only runs evaluation if training metrics are above a certain level. Adding a conditional component that only runs evaluation when training metrics exceed a certain threshold prevents unnecessary evaluation runs on poorly performing models. This reduces pipeline failures by ensuring that evaluation, which may be resource-intensive or prone to failure with low-quality inputs, is only triggered when the model has demonstrated sufficient training performance. This approach optimizes resource usage and pipeline reliability without sacrificing the evaluation step entirely.

Answer

Reduce the training dataset size

Answer

Remove the evaluation component

Answer

Increase the threshold value

Question 15

Refer to the exhibit. A data engineer sees these metrics from Cloud Monitoring for a deployed Vertex AI Endpoint. What is the most effective action to reduce latency?

Accepted Answer

Increase the number of replicas. The metrics show high CPU utilization and increasing latency, indicating the current instance is overloaded. Increasing the number of replicas distributes the inference requests across multiple instances, reducing per-replica load and lowering response times. This is the most direct way to scale horizontally and address latency caused by resource saturation.

Answer

Switch to batch prediction

Answer

Reduce the machine type

Answer

Enable model quantization