Google Cloud

Google Cloud Certified – Professional Machine Learning Engineer

PR000269
Trending

Validate your skills as a Professional Machine Learning Engineer with exam code PR000269 in Google Cloud.

495 questions 0 views Free
Start Mock Test Timed · Full-length · Scored

Questions 471–480 of 495

Q471

You are configuring IAM roles for structured data access. What happens if a user is granted a role with limited access to a resource but already has broader access due to another role?

  • A User loses broader access immediately
  • B User's access remains as per the broadest role
  • C Limited access overrides broader access
  • D User gains new permissions only
Explanation Google Cloud IAM uses the principle of least privilege, giving users the broadest access they have across roles.
Q472

Which service is used for real-time data processing in Google Cloud?

  • A Cloud Dataflow
  • B Cloud Storage
  • C Cloud Functions
  • D BigQuery
Explanation Cloud Dataflow is designed for real-time data processing, while the others serve different purposes.
Q473

A company needs to deploy a machine learning model with high availability. What should they consider?

  • A Use Cloud TPU for training
  • B Use Google Kubernetes Engine
  • C Deploy on Cloud Functions
  • D Use a single VM instance
Explanation Google Kubernetes Engine provides high availability through container orchestration, unlike the other options.
Q474

What happens when you enable Vertex AI's feature store for a project?

  • A Data is instantly stored in BigQuery
  • B All data becomes public
  • C You can version your ML data
  • D Training jobs are immediately deployed
Explanation Enabling Vertex AI's feature store allows you to version your ML data, not the other outcomes.
Q475

Which service is best for real-time data streaming?

  • A Cloud Pub/Sub
  • B Cloud SQL
  • C Cloud Storage
  • D BigQuery
Explanation Cloud Pub/Sub is designed for real-time streaming, while the others focus on storage or querying of data.
Q476

A company needs to train machine learning models with sensitive data. What should they use to manage access securely?

  • A Service Accounts
  • B Public IPs
  • C Object Versioning
  • D Firewall Rules
Explanation Service Accounts allow secure access management, whereas the other options do not address access control effectively.
Q477

You are configuring a model versioning system in AI Platform. What happens when you enable 'automatic rollback'?

  • A All model versions delete automatically
  • B Last failed version remains active
  • C Rolling back is manual process
  • D Latest version is always active
Explanation Enabling automatic rollback keeps the last successful version active after a failure, contrary to the incorrect options provided.
Q478

Which service is best for managing machine learning models in production?

  • A AI Platform
  • B BigQuery
  • C Cloud Functions
  • D Cloud Run
Explanation AI Platform is specifically designed for managing ML models, while others serve different primary purposes.
Q479

A company needs to process real-time streaming data. Which Google Cloud service should they use?

  • A Cloud SQL
  • B Cloud Pub/Sub
  • C Cloud Storage
  • D Firestore
Explanation Cloud Pub/Sub is designed for real-time messaging, enabling efficient data processing and ingestion.
Q480

What happens when you deploy a TensorFlow model with an unsupported TensorFlow version on AI Platform?

  • A Deployment succeeds with warnings
  • B Model fails to deploy
  • C Model deploys but is unstable
  • D Model is converted automatically
Explanation The model will fail to deploy if the version is unsupported, as compatibility is required.