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
Questions 371–380 of 495
A company needs to securely manage API keys for their services. What feature should they use in Google Cloud?
-
A
Cloud Storage
-
B
Secret Manager
-
C
Cloud Pub/Sub
-
D
Cloud Logging
Explanation
Secret Manager securely manages API keys; other options do not specifically target secret management.
You are configuring BigQuery permissions for a user. What happens if you grant them the 'roles/bigquery.jobUser' role?
-
A
They can create datasets.
-
B
They can run jobs only.
-
C
They can delete datasets.
-
D
They can update table schema.
Explanation
The 'roles/bigquery.jobUser' role allows users to run jobs only, not create or modify datasets or schemas.
Which service provides managed machine learning models in GCP?
-
A
AI Platform
-
B
Cloud Spanner
-
C
Firebase
-
D
BigQuery
Explanation
AI Platform is specifically designed for building and deploying ML models, while the others serve different purposes.
A company needs to process large-scale, real-time data streams. Which GCP service is best suited for this task?
-
A
Cloud Dataflow
-
B
Cloud Functions
-
C
Cloud Storage
-
D
BigQuery
Explanation
Cloud Dataflow is designed for real-time data processing, while the others do not specialize in stream processing.
What happens when you set a GCP resource to 'preemptible'?
-
A
Resource is always available
-
B
Resource can be terminated anytime
-
C
Resource has higher cost
-
D
Resource is immune to outages
Explanation
Preemptible resources can be terminated when GCP needs the capacity, while others do not have this characteristic.
Which Google Cloud service provides a fully managed NoSQL database?
-
A
Firestore
-
B
BigQuery
-
C
Cloud SQL
-
D
Cloud Storage
Explanation
Firestore is a NoSQL database service, while the others serve different database or storage purposes.
A company needs to process streaming data in real time. Which service should they use?
-
A
Cloud Functions
-
B
Pub/Sub
-
C
Dataproc
-
D
Cloud Run
Explanation
Pub/Sub is designed for real-time messaging and event-driven processing, unlike the other options.
You are configuring a Vertex AI model to predict outcomes. What happens if you set the batch size too high?
-
A
Model accuracy increases.
-
B
More memory usage may occur.
-
C
Prediction time decreases.
-
D
Results are returned faster.
Explanation
Setting a high batch size can lead to excessive memory usage, while the other options are incorrect effects.
Which service allows real-time data processing in GCP?
-
A
Dataflow
-
B
BigQuery
-
C
Cloud Storage
-
D
Pub/Sub
Explanation
Dataflow enables real-time data processing, while others serve different purposes.
A company needs to build a model that predicts customer churn based on multiple features. What approach should they take?
-
A
Use a clustering algorithm
-
B
Utilize regression analysis
-
C
Implement a reinforcement learning model
-
D
Apply natural language processing
Explanation
Regression analysis is suitable for predicting continuous outcomes like churn probability, whereas others are not applicable here.