Cloud Digital Leader Free Practice Questions — Page 9 (2026)

Question 81

Within Google’s Site Reliability Engineering framework, which concept measures how well a system is performing?

A. Service-level agreement

B. Service-level indicator

C. Error reporting

D. Service-level objective

Show Answer

Correct Answer: B

Explanation:
A Service-Level Indicator (SLI) is the quantitative metric used to measure how well a service is performing (such as latency, availability, error rate, or throughput). A Service-Level Objective (SLO) is the target value for an SLI, and a Service-Level Agreement (SLA) is the contractual commitment based on service levels. Error reporting is only one operational practice, not the core performance measurement concept.

Question 82

A cinema company wants to build a model to predict customer visit patterns for the coming year. They have three years of customer visit data across 300 theaters; however, the data has been stored in different formats by different theaters. They must train the ML model. What should they do?

A. Use the last year of data so there are fewer inconsistencies for the model to handle.

B. Transform the data into a consistent format.

C. Group different format types and train a different model for each group.

D. Choose an ML model type that can process different formats of input data.

Show Answer

Correct Answer: B

Explanation:
Machine learning models require consistent, standardized input data for effective training. The correct preprocessing step is to transform data from all theaters into a common schema and format before training. Using only one year wastes valuable historical data, training separate models by format adds unnecessary complexity, and selecting a model to handle inconsistent storage formats does not address the underlying data quality and schema consistency problem.

Question 83

An organization must identify and fix security vulnerabilities in its cloud infrastructure and applications. Which Google Cloud service should they use?

A. Security Command Center

B. Google Cloud Armor

C. Cloud Storage

D. VPC networks

Show Answer

Correct Answer: A

Explanation:
Security Command Center is Google Cloud's centralized security and risk management platform. It identifies vulnerabilities, misconfigurations, and security findings across Google Cloud resources and applications, helping organizations prioritize and remediate risks. Cloud Armor provides DDoS and WAF protection, Cloud Storage is an object storage service, and VPC networks provide networking capabilities rather than vulnerability management.

Question 84

An organization wants to leverage cloud technologies but is concerned about vendor lock-in. What would mitigate this concern?

A. Open standards

B. Database services

C. Service level agreements

D. Scalable infrastructure

Show Answer

Correct Answer: A

Explanation:
Open standards mitigate vendor lock-in by promoting interoperability and portability across cloud providers. Using standardized APIs, formats, and technologies reduces dependence on proprietary vendor-specific services, making migration easier. Database services, SLAs, and scalable infrastructure do not directly address vendor lock-in.

Question 85

An organization wants to use an open source library with a flexible ecosystem of tools to create and train its own machine learning models. Which product or solution should the organization use?

A. Cloud Functions

B. Apache Beam

C. Dataflow

D. TensorFlow

Show Answer

Correct Answer: D

Explanation:
TensorFlow is Google's open-source machine learning framework that provides a flexible ecosystem of tools, libraries, and community resources for building, training, and deploying machine learning models. Cloud Functions is a serverless compute service, Apache Beam is a unified programming model for data processing, and Dataflow is Google's managed data processing service based on Apache Beam.

Question 86

An organization has petabytes of data gathered from a wide range of sources. They want to use the data for strategic analysis and to guide business decisions. What type of service should they use?

A. Multi cloud environment

B. Virtual machine environment

C. Hybrid cloud environment

D. Container environment

Show Answer

Correct Answer: A

Explanation:
The question appears to have mismatched answer choices. For analyzing petabytes of data from many sources to support strategic decisions, the appropriate service would normally be a big data analytics platform or data lake, neither of which is listed. Of the available options, a multi-cloud environment is the closest fit because it can aggregate and analyze data across multiple cloud providers. Virtual machines, hybrid cloud, and containers are deployment models rather than analytics services, and containers in particular are not a data analytics service.

Question 87

What is an example of unstructured data?

A. Historical stock inventory

B. Product ratings

C. Customer orders

D. Call center transcripts

Show Answer

Correct Answer: D

Explanation:
Unstructured data does not conform to a predefined schema or tabular format. Call center transcripts are free-form text and are a classic example of unstructured data. Historical stock inventory, product ratings, and customer orders are typically stored in structured fields within databases.

Question 88

What is a defining feature of a non-relational database?

A. Reporting across multiple data sources

B. A strictly enforced schema

C. A flexible data model

D. Queries that join multiple tables

Show Answer

Correct Answer: C

Explanation:
A defining characteristic of non-relational (NoSQL) databases is a flexible data model with a schema that is often dynamic or schema-less, allowing storage of structured, semi-structured, or unstructured data. The other options describe characteristics more typical of relational databases, such as strict schemas and table joins.

Question 89

An organization needs to rapidly scale its use of computing resources and honor its commitment to environmental sustainability. What should the organization do?

A. Refactor application software to use less energy.

B. Use a public cloud provider with energy-efficient data centers.

C. Use a carbon-neutral energy provider for an existing on-premises data center.

D. Purchase energy-efficient servers for an existing on-premises data center.

Show Answer

Correct Answer: B

Explanation:
A public cloud provider with energy-efficient data centers allows the organization to rapidly scale computing resources on demand while benefiting from optimized infrastructure that is typically more energy-efficient than most on-premises deployments. The other options either do not address rapid scalability or require continuing to operate and expand an existing on-premises environment.

Question 90

A vacation home rental organization wants to predict the popularity of properties in their upcoming busy season. They do not have a data science team, and want to use their in-house database administration skills to create a machine learning model. What should the organization do?

A. Use custom training in Vertex AI with TensorFlow.

B. Integrate pre-trained APIs into their application.

C. Use BigQuery ML and create models using SQL.

D. Build a model in AutoML using labeled data.

Show Answer

Correct Answer: C

Explanation:
BigQuery ML is designed for users with SQL and database skills to build and train machine learning models directly in BigQuery without requiring a dedicated data science team or TensorFlow expertise. Pre-trained APIs are for existing AI capabilities rather than custom popularity prediction, and AutoML is no-code but does not specifically leverage the organization's existing database administration and SQL skills.