CDL Practice Questions
Prepare for CDL with more than an answer.
Unlock the full exam and previous versions
- v1Version 1 197 questions Current
- Cloud-Digital-LeaderLegacy Google Cloud Digital Leader 145 questions Locked
- Exam fee
- $99 USD
- Level
- Foundational
- Valid for
- 3 years
Domains covered on the exam 6
- Digital Transformation with Google Cloud17%
- Exploring Data Transformation with Google Cloud17%
- Innovating with Google Cloud Artificial Intelligence16%
- Modernizing Infrastructure and Applications with Google Cloud17%
- Trust and Security with Google Cloud17%
- Scaling with Google Cloud Operations16%
- 1
A company's finance department is struggling with unpredictable cloud bills. They need to implement controls to prevent project teams from exceeding their allocated quarterly budgets. Which two Google Cloud features are most effective for controlling and monitoring cloud spending? (Select TWO)
Show answer details
Correct answer: B, D
- 2
A large enterprise is considering a full migration to the cloud but is hesitant due to the perceived loss of control and the risks associated with new technologies. Which of the following represents a primary risk or implication for an organization that chooses not to adopt modern cloud technology?
Show answer details
Correct answer: C
Organizations that resist adopting cloud technology risk falling behind competitors who leverage the cloud for its agility, scalability, and advanced data/AI capabilities. The inability to quickly provision resources, experiment with new ideas, and analyze large datasets can lead to slower product development, poorer customer experiences, and ultimately, a loss of market share.
- 3
A marketing team wants to analyze customer sentiment by processing thousands of product reviews written in text form. They need a solution that can identify the overall sentiment (positive, negative, neutral) and extract key entities (like product names) without requiring any custom machine learning model development. Which Google Cloud service is the most direct and appropriate solution?
Show answer details
Correct answer: C
The Cloud Natural Language API is a pre-trained model that is perfect for this use case. It can perform sentiment analysis, entity extraction, and syntax analysis on text with a simple API call. Since the requirement is to use an existing solution without custom model development, this pre-trained API is the most efficient and direct choice.
- 4
True or False: Using a data warehouse is most appropriate for storing and analyzing large volumes of unstructured data like images, videos, and social media posts in their native format.
Show answer details
Correct answer: B
This statement is false. Data warehouses, like BigQuery, are optimized for storing and analyzing structured and semi-structured data. For storing vast quantities of unstructured data in its raw format, a data lake (often built on object storage like Google Cloud Storage) is the more appropriate solution.
- 5
A CTO is evaluating different levels of AI/ML solutions on Google Cloud for a new product recommendation engine. The three main approaches are using a pre-trained API, using AutoML, or building a custom model. What is the key trade-off the CTO must consider when choosing between these options?
graph LR A[Pre-trained API] --> B(Fastest Time-to-Market); A --> C(Least Differentiation); D[AutoML] --> E(Moderate Time-to-Market); D --> F(Some Differentiation); G[Custom Model] --> H(Slowest Time-to-Market); G --> I(Highest Differentiation);Show answer details
Correct answer: B
This is the fundamental trade-off. Pre-trained APIs are very fast to implement with low expertise required, but they offer little competitive differentiation as any competitor can use them. Building a custom model requires significant time, cost, and expertise but can create a unique, proprietary solution that provides a strong competitive advantage. AutoML sits in the middle, offering a balance of speed and customization.
- 6
A global retail company is adopting a hybrid cloud strategy. They need to manage their containerized applications consistently across their on-premises data centers and Google Cloud. A key requirement is a single management interface to observe, manage, and enforce policies for all their Kubernetes clusters, regardless of location. Which Google Cloud product is specifically designed to meet this need for unified, multi-environment container management?
Show answer details
Correct answer: C
GKE Enterprise (formerly known as Anthos) is the correct answer. It is designed to provide a consistent development and operations experience for cloud-native applications across hybrid and multi-cloud environments. It offers a single control plane for managing Kubernetes clusters, whether they are on Google Cloud, on-premises, or in other public clouds, which directly addresses the company's requirement for a unified management interface.
- 7
A financial services firm is planning its digital transformation. The executive team is concerned about the shift from a predictable, upfront capital expenditure (CapEx) model for IT to a variable operational expenditure (OpEx) model in the cloud. Which two statements accurately describe the business implications of this financial shift? (Select TWO)
Show answer details
Correct answer: A, C
- 8
A startup is developing a mobile application that allows users to identify plant species by uploading a photo. The development team has no machine learning expertise but has collected a large, high-quality dataset of labeled plant images. They need to create a custom model that is highly accurate for their specific task. Which Google Cloud AI/ML solution provides the best balance of custom model creation, high performance, and minimal required ML expertise?
Show answer details
Correct answer: C
AutoML Vision is the ideal solution. It allows teams with limited ML expertise to train high-quality, custom image classification models using their own labeled data. This directly addresses the startup's need to create a specialized model for plant species identification without requiring them to build and manage the underlying ML infrastructure from scratch. The pre-trained Vision API is too generic, and building a custom TensorFlow model requires deep ML expertise, which the team lacks.
- 9
A university is setting up its Google Cloud organization. It needs to provide distinct levels of access and apply different policies for various departments. The 'Research' department needs access to powerful, expensive resources, while the 'Administration' department has standard requirements, and 'Student Labs' must be heavily restricted to prevent accidental overspending. What is the most effective and scalable way to structure their Google Cloud environment to enforce these differing requirements?
Show answer details
Correct answer: C
Using the resource hierarchy with folders is the best practice for this scenario. Creating a folder for each department allows administrators to apply distinct IAM permissions, Organization Policies (e.g., restricting VM types for students), and budgets that are inherited by all projects within that folder. This provides centralized control, scalability, and clear separation of concerns, which is far more effective than managing policies project-by-project or tracking costs with labels alone.
graph TD Org(Organization: University.edu) --> F_Research(Folder: Research) Org --> F_Admin(Folder: Administration) Org --> F_Labs(Folder: Student Labs) F_Research --> P_HPC(Project: HPC Study) F_Admin --> P_Finance(Project: Finance App) F_Labs --> P_CS101(Project: CS101 Lab) style F_Research fill:#f9f,stroke:#333,stroke-width:2px style F_Admin fill:#ccf,stroke:#333,stroke-width:2px style F_Labs fill:#cfc,stroke:#333,stroke-width:2px - 10
A media company processes large video files. The workflow involves an initial, frequent access period of 30 days for editing, followed by a 90-day period where the files might be accessed once for quality control, and finally, long-term archival for 7 years with access being extremely rare. To optimize storage costs, which combination of Cloud Storage classes should be used for each phase of the data lifecycle?
Show answer details
Correct answer: A
This combination correctly maps the access patterns to the most cost-effective storage classes. Standard is for frequently accessed ('hot') data. Nearline is ideal for data accessed infrequently (e.g., once a month), fitting the 90-day quality control period. Archive is the lowest-cost option designed for long-term data archival where access is rare, perfectly matching the 7-year requirement.
