Courseiva

Google Cloud Digital Leader (GCDL) — Questions 376–450

848 questions total · 12 pages · All types, answers revealed

Page 5

Page 6 of 12

Page 7
376
MCQhard

A DevOps engineer wants to automatically delete a Cloud Storage object after 30 days from creation. The object is stored in a bucket with the 'Standard' storage class. Which approach should the engineer use?

A.Set a retention policy on the bucket with a 30-day retention period
B.Change the storage class to 'Archive' which automatically deletes after 30 days
C.Enable object versioning and set a noncurrent time deletion of 30 days
D.Add a lifecycle rule to delete objects after 30 days
AnswerD

A bucket lifecycle rule supports an age-based Delete action, which automatically removes objects after a specified number of days from their creation time. By setting 'age: 30' in the Delete action, Cloud Storage will delete the objects exactly at the 30-day mark during its daily lifecycle evaluation. This is the native, recommended mechanism for automatic object expiration and directly fulfills the requirement.

Why this answer

A lifecycle rule in Cloud Storage allows automatic deletion of objects after a specified age, such as 30 days from creation. This is the standard method to manage object lifecycle and reduce storage costs. The rule can be applied to a bucket and will delete objects matching the criteria.

Exam trap

GCDL often tests the confusion between retention policies (which prevent deletion) and lifecycle rules (which perform deletion), leading candidates to choose retention when deletion is required.

How to eliminate wrong answers

Option A is wrong because a retention policy prevents deletion until the retention period expires; it does not delete objects. Option B is wrong because changing storage class to Archive does not automatically delete objects; it only changes storage class for cost savings. Option C is wrong because object versioning with noncurrent time deletion only deletes noncurrent versions after a period, not the live object.

377
MCQhard

A financial services company is migrating its on-premises monitoring system to Google Cloud. They need to collect metrics, logs, and traces from multiple projects and provide a unified view for their operations team. Security requires that logs containing sensitive data be stored with additional encryption and access controls. Which combination of services should they use?

A.Cloud Monitoring, Cloud Logging, and Cloud Trace with Logging's _Required and _Default buckets.
B.Cloud Monitoring, Cloud Logging, and Cloud Trace with a custom sink to a BigQuery dataset that uses CMEK.
C.Cloud Monitoring and Cloud Logging with Log Analytics.
D.Cloud Monitoring, Cloud Logging, and Cloud Trace with Cloud Audit Logs.
AnswerB

A custom log sink routes selected log entries to a user-controlled destination, here a BigQuery dataset encrypted with CMEK, enabling the customer to manage and rotate the encryption keys via Cloud KMS. Pairing this with Cloud Monitoring for resource metrics and Cloud Trace for distributed request spanning gives complete observability: logs, metrics, and traces are unified. The CMEK-protected BigQuery dataset also allows fine-grained IAM access control and retention management, satisfying both encryption and least-privilege requirements.

Why this answer

The company needs to collect metrics, logs, and traces (requiring Cloud Monitoring, Cloud Logging, and Cloud Trace) and must store logs containing sensitive data with additional encryption and access controls. A custom sink to BigQuery with CMEK provides customer-managed encryption keys for the BigQuery dataset, and BigQuery's native access controls (IAM, row-level security) satisfy the requirement for additional access controls beyond the default Logging buckets.

Exam trap

Google Cloud often tests the misconception that the _Required and _Default buckets are sufficient for compliance, but they lack CMEK and granular access controls, which are essential for sensitive data handling.

How to eliminate wrong answers

Option A is wrong because the _Required and _Default buckets are built-in Logging storage buckets that use Google-managed encryption keys (GMEK) by default and do not provide the additional encryption (CMEK) or granular access controls required for sensitive data. Option C is wrong because it omits Cloud Trace entirely, which is needed for collecting traces, and Log Analytics alone does not provide the separate, encrypted storage with custom access controls for sensitive logs. Option D is wrong because Cloud Audit Logs are a specific type of log (administrative activity, data access, etc.) and not a storage or encryption mechanism; they do not enable CMEK or custom access controls for sensitive data.

378
MCQeasy

A developer needs to debug a production issue by analyzing logs from multiple microservices. Which Google Cloud service should they use to filter and search logs in real time?

A.Cloud Monitoring
B.Error Reporting
C.Cloud Logging
D.Cloud Debugger
AnswerC

Cloud Logging is the correct service because it is purpose-built for ingesting, storing, searching, and analyzing logs in real time. It offers a powerful query language, filters, and the ability to view logs from a single VM, container, or Kubernetes cluster, making it the ideal tool for debugging a production issue. Cloud Logging also integrates with Cloud Monitoring and Error Reporting, but it alone provides the comprehensive log analysis functionality described in the scenario.

Why this answer

Cloud Logging (formerly Stackdriver Logging) is the correct service because it provides a centralized log management system that can ingest logs from multiple microservices, filter them using advanced queries, and search them in real time. Its Logs Explorer interface supports custom filters, labels, and timestamps, enabling developers to pinpoint production issues across distributed services without delay.

Exam trap

Google Cloud often tests the distinction between log management (Cloud Logging) and error aggregation (Error Reporting), leading candidates to choose Error Reporting when the question explicitly asks for filtering and searching logs in real time.

How to eliminate wrong answers

Option A is wrong because Cloud Monitoring focuses on metrics, uptime checks, and alerting policies, not on filtering or searching raw log data in real time. Option B is wrong because Error Reporting automatically aggregates and analyzes application errors (e.g., stack traces) but does not provide a general-purpose log search or filtering capability for arbitrary log entries. Option D is wrong because Cloud Debugger allows you to inspect the state of a running application (e.g., capture snapshots and logpoints) without stopping it, but it is not designed for centralized log aggregation, filtering, or real-time search across multiple microservices.

379
MCQmedium

A company's compliance team asks what evidence they can provide to regulators to demonstrate that Google Cloud services meet industry security standards. Which type of documentation most directly provides this evidence?

A.Google Cloud's marketing materials and product documentation describing security features
B.Third-party audit reports and compliance certifications (SOC 2 Type II, ISO 27001, PCI DSS) available through Google Cloud's Compliance Reports Manager, which provide independent verification of security controls
C.The company's own internal security policies that reference using Google Cloud
D.A Google Cloud support ticket confirming that the company's account is in good standing
AnswerB

These reports are the correct evidence. SOC 2 Type II demonstrates security controls operated effectively over a period. ISO 27001 certification shows a comprehensive ISMS is in place. PCI DSS attestation covers payment card security. These are issued by qualified independent auditors and accepted by regulators globally.

Why this answer

Third-party audit reports and compliance certifications (SOC 2 Type II, ISO 27001, PCI DSS) available through Google Cloud's Compliance Reports Manager provide independent verification of security controls. These are the most direct evidence for regulators because they are produced by independent auditors and attest to the effectiveness of controls.

Exam trap

GCDL often tests the confusion between self-reported security features and independently audited compliance reports, leading candidates to choose marketing materials or internal policies as evidence.

How to eliminate wrong answers

Option A is wrong because marketing materials and product documentation are not independent evidence; they are self-reported and not audited. Option C is wrong because internal security policies are the company's own documents, not evidence of Google Cloud's compliance. Option D is wrong because a support ticket confirming account standing does not provide any security control verification.

380
MCQmedium

A large hospital network wants to move patient records to the cloud and enable doctors to access records from any device. The Chief Medical Officer is supportive, but the legal department raises data privacy concerns, and the IT department fears job losses. Which aspect of digital transformation does this scenario highlight?

A.The primary challenge is selecting the correct cloud database for patient records.
B.Digital transformation requires aligning technology with people and culture — managing stakeholder concerns and change resistance is often harder than the technical migration.
C.The hospital should delay cloud adoption until quantum computing makes it more secure.
D.The legal department's concerns prove that healthcare organizations cannot use public cloud.
AnswerB

The correct answer centers on sociotechnical transformation: technology is the enabler, but the real hard part is people. Legal fears about compliance, clinicians' concerns about workflow disruption, and IT's operational anxiety are all human/organizational barriers. Successful cloud adoption in healthcare requires structured change management, transparent communication, and meaningful stakeholder engagement to convert resistance into co-ownership.

Why this answer

Digital transformation is not solely about technology adoption; it critically involves managing the human and cultural aspects of change. In this scenario, the legal department's privacy concerns and the IT department's fear of job losses represent stakeholder resistance that must be addressed through communication, retraining, and policy alignment. Successful cloud migration in healthcare requires balancing technical migration with change management to ensure adoption and compliance.

Exam trap

Google Cloud often tests the misconception that digital transformation is purely a technology challenge, leading candidates to focus on technical solutions (like database selection or security improvements) rather than recognizing that people and culture are the harder, more critical components of successful transformation.

How to eliminate wrong answers

Option A is wrong because the primary challenge is not selecting the correct cloud database; while database choice is a technical consideration, the scenario explicitly highlights stakeholder concerns (legal and IT) as the core issue, not a technical selection problem. Option C is wrong because delaying cloud adoption for quantum computing is impractical and unnecessary; current cloud security measures like encryption at rest (AES-256) and in transit (TLS 1.2/1.3), along with HIPAA-compliant configurations, already provide adequate protection for patient records. Option D is wrong because the legal department's concerns do not prove that healthcare organizations cannot use public cloud; many healthcare providers successfully use public cloud platforms (e.g., AWS, Azure, GCP) with proper compliance frameworks like HIPAA BAA, access controls, and data residency policies.

381
MCQeasy

According to the shared responsibility model, which of the following is the customer responsible for?

A.Physical security of data centers
B.Network infrastructure security
C.Data encryption and access control (IAM)
D.Hypervisor security
AnswerC

Customers retain responsibility for encrypting their data both at rest and in transit, as well as for defining and enforcing access control through IAM policies. This includes managing encryption keys (via Cloud KMS or customer-supplied keys), configuring identity and role-based access, and applying organizational policies. Since these are direct customer actions within the cloud console and APIs, this is the correct choice.

Why this answer

Under the shared responsibility model, customers are responsible for securing their data and controlling access to it. This includes encrypting data (at rest and in transit, where applicable) and managing identity and access with IAM policies. Google secures the physical and foundational infrastructure but does not manage customer IAM or data encryption choices.

Exam trap

GCDL often tests the misconception that Google handles encryption and access control by default — candidates must remember that IAM and data encryption configuration are customer responsibilities.

How to eliminate wrong answers

Option A is wrong because physical security of data centers is Google's responsibility, not the customer's. Option B is wrong because network infrastructure security (the underlying Google network) is managed by Google. Option D is wrong because hypervisor security is part of Google's managed infrastructure layer, not the customer's responsibility.

382
MCQmedium

A startup founder argues that her company has an advantage over established enterprises when adopting cloud-native technologies. Which characteristic of startups most supports this claim in the context of digital transformation?

A.Startups have larger technology budgets than enterprises, allowing them to purchase more cloud services
B.Startups have no legacy systems or organizational inertia, allowing them to build cloud-native from day one without migration complexity
C.Cloud providers offer preferential pricing to startups, giving them a cost advantage over enterprises
D.Startups employ more skilled engineers than enterprises because they offer higher salaries
AnswerB

This is the core startup advantage in digital transformation: a greenfield environment. No legacy systems to integrate, no entrenched processes to change, no organizational inertia to overcome. Cloud-native architecture can be adopted from the first line of code.

Why this answer

Startups lack legacy systems and organizational inertia, which are the primary barriers to adopting cloud-native architectures. Established enterprises often face complex migration challenges, technical debt, and rigid processes that slow digital transformation. By building cloud-native from day one, startups can leverage microservices, containers, and serverless computing without the cost and risk of re-architecting existing systems.

Exam trap

Google Cloud often tests the misconception that cost or budget is the primary driver of cloud adoption, when in reality the absence of legacy technical debt and organizational inertia is the decisive factor for startups in digital transformation.

How to eliminate wrong answers

Option A is wrong because startups typically have smaller technology budgets than established enterprises, not larger; cloud-native adoption is driven by agility and lack of legacy constraints, not by spending capacity. Option C is wrong because while some cloud providers offer startup credits, this is a temporary financial incentive and does not address the fundamental advantage of avoiding migration complexity and legacy dependencies. Option D is wrong because startups generally cannot match enterprise salaries and often have fewer engineers; the advantage lies in organizational flexibility, not in hiring more skilled personnel.

383
MCQeasy

A company stores customer data in Google Cloud and wants to ensure data confidentiality in the event that hardware is decommissioned and returned by Google. How does Google protect customer data when storage hardware reaches end of life?

A.Google transfers customer data to new hardware first, then ships the old hardware to the customer for self-destruction.
B.Google uses approved data erasure and physical destruction processes (shredding, degaussing) for decommissioned storage media before hardware leaves its facilities.
C.Customer data on decommissioned hardware is automatically encrypted, making it safe to discard without wiping.
D.Customers must pay a data destruction fee to ensure their data is wiped from decommissioned hardware.
AnswerB

This is the correct answer. Google follows NIST 800-88-compliant sanitization procedures for storage media, combining cryptographic erase (secure key destruction) with physical destruction methods such as shredding and degaussing. These steps are performed inside Google's data centers before any hardware leaves the premises, and the process is independently audited to verify effectiveness. This guarantees that customer data cannot be reconstructed from decommissioned media.

Why this answer

Google Cloud follows strict data destruction policies for decommissioned storage media. Before any hardware leaves Google's facilities, it undergoes approved data erasure (e.g., NIST SP 800-88 compliant wiping) followed by physical destruction (e.g., shredding, degaussing) to ensure customer data cannot be recovered. This process guarantees data confidentiality even if the hardware is returned or recycled.

Exam trap

The trap here is that candidates assume encryption alone (Option C) is sufficient for decommissioned hardware, but Google's policy requires physical destruction or verified erasure to prevent data recovery from encrypted drives if keys are later compromised.

How to eliminate wrong answers

Option A is wrong because Google does not ship decommissioned hardware to customers; instead, Google retains and destroys the hardware internally to prevent any data leakage. Option C is wrong because while data at rest is encrypted, encryption alone is not sufficient for decommissioned hardware—Google still performs secure erasure and physical destruction to protect against future decryption or key compromise. Option D is wrong because data destruction is included as a standard part of Google's hardware lifecycle management at no additional cost to customers.

384
MCQmedium

A development team runs a web application on Cloud Run. They need to store session state across requests. Which Google Cloud service should they use?

A.Memorystore for Redis
B.Cloud Storage
C.Cloud SQL
D.Cloud Pub/Sub
AnswerA

Memorystore for Redis is a fully managed in-memory data store compatible with the Redis protocol, offering sub-millisecond latency for key-value lookups—exactly what session state requires. Cloud Run instances are stateless and ephemeral, scaling to zero when idle, so placing sessions in a shared cache like Memorystore ensures user state survives individual instance lifecycles and remains accessible across autoscaled replicas. Furthermore, Redis-native features such as TTLs, atomic increments, and SETNX are ideal for session expiry, rotation, and concurrency control, making it the correct choice for low-latency session persistence.

Why this answer

Cloud Run instances are stateless; for session state, use an external caching layer like Memorystore (Redis). Pub/Sub is for messaging, Cloud SQL for relational data, and Cloud Storage for objects.

385
MCQhard

A company's cloud cost has grown significantly. A FinOps analysis reveals the largest waste category is idle Cloud SQL instances — 12 database instances that were provisioned for projects that have since ended, but were never deleted. What process failure most directly caused this waste?

A.The company should have used a cheaper database service instead of Cloud SQL
B.The absence of a resource decommissioning process: when projects end, there is no formal step to identify and delete associated cloud resources, allowing idle infrastructure to persist and accrue costs indefinitely
C.Cloud SQL pricing is too high compared to on-premises databases, making any unused capacity expensive
D.The database administrators forgot to enable automatic deletion for idle Cloud SQL instances
AnswerB

This is the root cause. FinOps best practice requires a defined lifecycle process: when a project is closed or a service is decommissioned, associated cloud resources are explicitly identified and deleted. Without this step, idle resources accumulate. The fix is process: add resource cleanup to the project closure checklist and automate detection of idle resources.

Why this answer

The root cause is the lack of a formal resource decommissioning process. When projects end, there is no automated or manual step to identify and delete associated Cloud SQL instances, so idle databases continue to incur costs. In Google Cloud, Cloud SQL instances do not auto-delete; they persist until explicitly removed, making a decommissioning workflow essential to prevent waste.

Exam trap

The GCDL exam often tests the concept that cloud resources are not automatically cleaned up when projects end, and candidates mistakenly think technical features like auto-deletion or cheaper services are the solution, rather than recognizing the need for a process-driven decommissioning workflow.

How to eliminate wrong answers

Option A is wrong because the waste is not due to the choice of database service; Cloud SQL is appropriate for relational workloads, and the issue is that instances are idle, not that a cheaper service would solve the problem of forgotten resources. Option C is wrong because comparing Cloud SQL pricing to on-premises databases is irrelevant; the waste is from unused capacity, not from the pricing model itself. Option D is wrong because Cloud SQL does not have an 'automatic deletion' feature for idle instances; the responsibility lies with the organization to implement lifecycle management, not with a missing configuration toggle.

386
MCQmedium

A company wants to migrate its on-premises PostgreSQL database to Google Cloud with minimal application changes. They need high availability and want to leverage AI-powered optimizations for performance. Which service should they choose?

A.Bigtable
B.Cloud SQL for PostgreSQL
C.AlloyDB
D.Cloud Spanner
AnswerC

AlloyDB is purpose-built for PostgreSQL workloads and is fully PostgreSQL-compatible, enabling most applications to migrate without changing code or SQL syntax. It delivers enterprise-grade high availability and performance with a columnar engine that speeds up analytical queries, adaptive indexes that learn from access patterns, and AI-driven optimization features that automatically fine-tune database settings—capabilities that Cloud SQL and other managed PostgreSQL services lack, making it the correct answer for this scenario.

Why this answer

AlloyDB for PostgreSQL is a fully managed, PostgreSQL-compatible database service on Google Cloud that offers high availability and uses AI-powered optimizations for performance, such as adaptive caching and automated tuning. It is designed for minimal application changes because it is wire-compatible with PostgreSQL. This makes it the best fit for migrating an on-premises PostgreSQL database with high availability and AI-driven performance.

Exam trap

GCDL often tests the confusion between Cloud SQL and AlloyDB, where candidates may choose Cloud SQL for PostgreSQL due to familiarity, missing the AI-powered optimizations and higher performance of AlloyDB.

How to eliminate wrong answers

Option A is wrong because Bigtable is a NoSQL wide-column database, not compatible with PostgreSQL, and would require significant application rewrites. Option B is wrong because Cloud SQL for PostgreSQL is a managed service but does not include AI-powered optimizations like AlloyDB; it is more suitable for smaller workloads. Option D is wrong because Cloud Spanner is a globally distributed relational database that is not PostgreSQL-compatible (though it has a PostgreSQL interface, it requires changes and is not a drop-in replacement).

387
MCQeasy

A company wants to run containerized applications on Google Cloud without managing the underlying infrastructure. Which service should they use?

A.Cloud Run
B.App Engine Flexible Environment
C.Compute Engine
D.Google Kubernetes Engine (GKE)
AnswerA

Cloud Run is the correct choice because it runs containerized applications on a fully managed serverless platform. You simply deploy a standard OCI container image, and Cloud Run automatically handles provisioning, scaling (including scaling to zero), and load balancing with no infrastructure to manage. It abstracts away all underlying compute and cluster operations, making it the simplest way to run stateless containers in production on Google Cloud.

Why this answer

Google Kubernetes Engine (GKE) is the managed Kubernetes service that abstracts node management, but it still requires some cluster management. Cloud Run is serverless and fully manages infrastructure, so it is the best choice for zero infrastructure management.

388
Multi-Selectmedium

Which TWO statements accurately describe Google Cloud's sustainability commitments? (Choose 2)

Select 2 answers
A.All Google Cloud regions are already carbon-neutral.
B.Google Cloud offsets all of its historical carbon emissions.
C.Google Cloud aims to run on 24/7 carbon-free energy in all its data centers by 2030.
D.Google Cloud matches 100% of its global electricity consumption with renewable energy.
E.Google Cloud data centers are powered exclusively by on-site solar panels.
AnswersC, D

This statement correctly identifies Google's public commitment to achieve 24/7 carbon-free energy across all its data centers and cloud regions by 2030. This target exceeds annual renewable matching, requiring that every hour of the day be served by carbon-free sources, which drives investments in local wind, solar, and storage. It is a time-matched, location-specific goal that differentiates Google from companies that only procure renewables on an annual basis.

Why this answer

Option C is correct because Google has committed to operating all of its data centers on 24/7 carbon-free energy (CFE) by 2030, meaning every hour of every day will be matched by clean energy sources on the same regional grid. Option D is correct because Google has matched 100% of its global annual electricity consumption with renewable energy purchases since 2017, a commitment that applies across its global operations including Google Cloud. Option A is not accurate as stated because carbon neutrality applies to Google's overall operations through offsets and renewable matching, not to every individual cloud region being inherently carbon-neutral.

Option B is incorrect because Google's commitment is to offset its historical carbon emissions cumulatively by 2030, not that it has already offset all historical emissions. Option E is incorrect because Google's data centers are not powered exclusively by on-site solar panels; they rely on a mix of grid electricity and renewable energy purchases such as PPAs.

Exam trap

GCDL often tests the distinction between annual renewable energy matching (100% since 2017) and 24/7 carbon-free energy (targeted by 2030), as well as the misconception that carbon neutrality means zero emissions or that all regions are already carbon-free.

389
MCQhard

A finance team needs to analyze detailed cost data across multiple projects, including resource-level breakdowns, labels, and cost attribution. They want to export this data to a BigQuery dataset for custom analysis. Which billing export option should they enable?

A.Use the Cloud Billing API to pull cost data programmatically
B.Set up a budget alert with notifications to Pub/Sub and ingest into BigQuery
C.Enable billing export to Cloud Storage in CSV format
D.Enable billing export to BigQuery (standard usage cost data)
AnswerD

This native integration streams detailed billing line items into a BigQuery dataset automatically, with daily and monthly tables that include metadata like resource labels, SKU, cost, and usage amount. It supports standard SQL queries for custom reports, joins with other datasets, and integrates with Looker Studio or other BI tools. The export is fully managed, eliminating custom ETL, and is the recommended approach for comprehensive cost analytics.

Why this answer

Enabling billing export to BigQuery (standard usage cost data) provides detailed, resource-level cost information including labels, projects, and SKUs, which can be queried directly in BigQuery for custom analysis. This is the only option that natively exports granular cost data to BigQuery without additional pipeline development. It supports cost attribution and detailed breakdowns required by the finance team.

Exam trap

The trap is assuming that budget alerts or Cloud Billing API can serve as a billing export to BigQuery. Candidates may think budget alerts provide detailed cost data, but they only trigger notifications based on thresholds.

How to eliminate wrong answers

Option A is wrong because the Cloud Billing API provides programmatic access to cost data but does not automatically export it to BigQuery; it would require custom code to pull and load data, adding complexity. Option B is wrong because budget alerts only notify when spending exceeds thresholds; they do not provide detailed cost data and are not a billing export mechanism. Option C is wrong because exporting to Cloud Storage in CSV format provides raw files but not the structured, queryable dataset in BigQuery that the team needs; it would require additional ETL to load into BigQuery.

390
Multi-Selectmedium

A company is deploying a global web application with static content (images, CSS, JS) and dynamic API calls. They want to reduce latency for users worldwide. Which TWO services should they combine? (Choose 2)

Select 2 answers
A.Cloud VPN
B.Cloud CDN
C.External HTTP(S) Load Balancer
D.Cloud Armor
E.Cloud DNS
AnswersB, C

Cloud CDN leverages Google's global edge PoPs to cache static responses (images, CSS, JS) close to end users, dramatically cutting latency and reducing origin load. It integrates with External HTTP(S) Load Balancing to honor cache-control headers and TTLs, making it the direct service for low-latency static content delivery.

Why this answer

Cloud CDN (option B) caches static content (images, CSS, JS) at Google's globally distributed edge points of presence (PoPs), reducing latency for users by serving content from a location closer to them. The External HTTP(S) Load Balancer (option C) is required to front the application and integrate with Cloud CDN, providing global anycast IP, SSL termination, and traffic distribution to backend instances. Together, they enable low-latency delivery of both static and dynamic content by combining edge caching with intelligent load balancing.

Exam trap

The trap here is that candidates often confuse Cloud CDN with Cloud Armor or Cloud DNS, thinking security or DNS services can reduce latency, but only CDN combined with a global load balancer provides edge caching and anycast routing for static and dynamic content.

391
MCQmedium

An e-commerce company needs a globally distributed relational database with strong consistency and 99.999% SLA to handle customer orders and inventory across multiple regions. They require SQL compatibility and automatic replication. Which database should they use?

A.Cloud Spanner
B.Bigtable
C.Cloud SQL
D.Firestore
AnswerA

Cloud Spanner is a fully managed relational database that combines the semantics of traditional SQL with horizontal scaling across regions. Using TrueTime, it provides external consistency (strong global consistency) while replicating data across continents, and it offers a 99.999% multi-region SLA. For an e-commerce platform needing global transactions on relational tables, Spanner uniquely satisfies both relational constraints and global distribution.

Why this answer

Cloud Spanner is Google Cloud's globally distributed, horizontally scalable relational database that provides strong consistency (external consistency via TrueTime) and a 99.999% availability SLA for multi-region configurations. It supports ANSI SQL and automatic synchronous replication across regions, which matches the e-commerce requirement for SQL compatibility, global distribution, and strong consistency for orders and inventory. No other option combines relational SQL, global scale, strong consistency, and a five-nines SLA.

Exam trap

GCDL often tests the trade-off between relational consistency and global scale — the trap is selecting Cloud SQL because it is 'SQL,' ignoring that it cannot deliver multi-region strong consistency or the 99.999% SLA.

How to eliminate wrong answers

Option B is wrong because Bigtable is a wide-column NoSQL database with eventual consistency and no SQL support — it is designed for high-throughput analytical and time-series workloads, not relational transactions. Option C is wrong because Cloud SQL is a regional (or HA within a region) managed MySQL/PostgreSQL/SQL Server instance; it does not provide global multi-region strong consistency or a 99.999% SLA. Option D is wrong because Firestore is a document NoSQL database with eventual consistency in multi-region mode and no relational SQL or multi-row ACID transactions across arbitrary queries at global scale.

392
MCQmedium

A company wants to implement a zero-trust security model to replace its legacy VPN for accessing internal web applications. Employees use both company-managed and personal devices. Which Google Cloud service provides context-aware access based on user identity and device posture?

A.Cloud VPN
B.Identity-Aware Proxy (IAP)
C.Cloud Armor
D.VPC Service Controls
AnswerB

Identity-Aware Proxy (IAP) is a Google Cloud service that sits in front of your applications and proxies every request, enforcing access control based on the requester's identity, group membership, and contextual attributes such as IP address, device security status, and geolocation. Because it operates at the application layer (HTTP/S or TCP) rather than the network layer, it can grant or deny access to individual resources without changing your network topology or requiring a VPN. IAP is the core enforcement point for a BeyondCorp-style zero-trust architecture, enabling context-aware, least-privilege access to both web applications and SSH/RDP services.

Why this answer

Identity-Aware Proxy (IAP) verifies user identity and context (device security, location, etc.) before granting access to applications, enabling zero-trust without a VPN.

393
MCQmedium

A company is migrating its on-premises data warehouse to Google Cloud. The data warehouse contains structured relational data (sales transactions, customer records) and is queried using standard SQL. The company wants to separate compute and storage costs and automatically scale compute resources up and down based on query load. Which Google Cloud service is most appropriate?

A.Cloud SQL
B.BigQuery
C.Dataproc
D.Cloud Spanner
AnswerB

BigQuery's separation of storage and compute lets slots scale independently of data at rest, satisfying the automatic scaling constraint. Its columnar, SQL-native engine handles structured relational sales and customer data without provisioning servers, unlike Cloud SQL or Dataproc, which couple compute to storage or require cluster management.

Why this answer

BigQuery is a serverless data warehouse that separates compute and storage, scales automatically, and supports standard SQL. Cloud SQL is for OLTP, not data warehousing. Spanner is globally distributed OLTP.

Dataproc is for Hadoop/Spark workloads, not directly for SQL warehousing.

394
MCQmedium

A security architect wants to implement a 'never trust, always verify' security approach where no user or service is assumed to be trustworthy based on network location alone. Every access request must be authenticated and authorized regardless of whether it comes from inside or outside the corporate network. Which security model describes this approach?

A.Perimeter security model
B.Zero Trust security model
C.Defense in depth model
D.Principle of least privilege
AnswerB

The Zero Trust security model fundamentally eliminates implicit trust based on network location, mandating continuous authentication, authorization, and encryption for every request, regardless of whether the source is inside or outside the corporate boundary. It operationalizes the 'never trust, always verify' principle through identity-centric policies, device posture checks, and microsegmentation, ensuring that even compromised internal hosts cannot move laterally without re-verification.

Why this answer

The Zero Trust security model (Option B) is correct because it explicitly enforces the 'never trust, always verify' principle, requiring authentication and authorization for every access request regardless of network location. In Google Cloud, this aligns with BeyondCorp, which uses identity-aware proxy (IAP) and context-aware access to verify each request based on user identity, device posture, and other attributes, rather than trusting based on IP address or network perimeter.

Exam trap

The trap here is that candidates often confuse 'defense in depth' (Option C) with Zero Trust because both involve multiple security layers, but defense in depth does not inherently require every request to be verified regardless of network location, which is the defining characteristic of Zero Trust.

How to eliminate wrong answers

Option A is wrong because the perimeter security model assumes trust inside the corporate network (e.g., VPN or firewall boundaries), which violates the 'never trust, always verify' approach. Option C is wrong because defense in depth is a layered security strategy (e.g., firewalls, IDS, encryption) but does not inherently require every request to be authenticated and authorized regardless of network location; it can still rely on perimeter trust. Option D is wrong because the principle of least privilege limits permissions to the minimum necessary but does not address the core requirement of verifying every access request based on location independence; it is a complementary concept, not the model described.

395
MCQmedium

An engineer needs to increase the default limit of 24 CPUs per region for Compute Engine instances in their project. They have already consumed 20 CPUs and need to launch a new instance with 8 CPUs. What should they do?

A.Launch the instance anyway; Google Cloud will automatically increase the quota.
B.Request a quota increase in the Cloud Console Quotas page.
C.Create a new project and launch the instance there.
D.Use a different region with available quota.
AnswerB

In the Cloud Console, navigate to IAM & Admin → Quotas, select the specific CPU quota (e.g., CPUs per region), click Edit, and submit a quota increase request. The request goes through a review and approval process; once approved, the new limit becomes effective in minutes to hours. This is the correct way to raise the default 24 CPU limit so you can launch your instance in the desired project and region.

Why this answer

Service quotas are per-project and per-region limits. The engineer must request a quota increase via the Quotas page in the Cloud Console before exceeding the limit.

396
MCQeasy

A global e-commerce company wants to build a product recommendation engine that suggests items to customers based on their real-time browsing behavior and purchase history. They want a pre-built solution that doesn't require building an ML recommendation model from scratch. Which Google Cloud product is purpose-built for retail recommendations?

A.BigQuery ML — build a collaborative filtering model using SQL.
B.Recommendations AI (Vertex AI Search for Retail)
C.Cloud SQL — query purchase history to find commonly bought-together products.
D.Cloud Dataflow — stream user clickstream data to build recommendations in real time.
AnswerB

Recommendations AI, now part of Vertex AI Search for Retail, is purpose-built for e-commerce personalization: it starts with deep neural network models pre-trained on large-scale retail patterns and fine-tunes them on your product catalog and user event data (views, clicks, purchases). The service automatically handles feature engineering, model training, and low-latency serving, returning personalized recommendations through a simple predict API. It also addresses cold-start users/items and supports optimizing for business objectives like revenue lift, without the need for ML engineering.

Why this answer

Recommendations AI (now part of Vertex AI Search for Retail) is Google Cloud's purpose-built, pre-built solution for retail product recommendations. It uses deep learning models trained on retail-specific data (e.g., clickstream, purchase history) to generate personalized suggestions without requiring the user to build or train an ML model from scratch.

Exam trap

The trap here is that candidates may confuse a general-purpose data/ML tool (like BigQuery ML, Cloud Dataflow, or Cloud SQL) with a purpose-built, pre-built solution for a specific domain (retail recommendations), leading them to choose an option that requires significant custom development instead of the turnkey service.

How to eliminate wrong answers

Option A is wrong because BigQuery ML requires you to write SQL to build and train a collaborative filtering model yourself, which is not a pre-built solution. Option C is wrong because Cloud SQL is a managed relational database service, not a recommendation engine; querying purchase history for co-purchased products would require custom application logic and does not provide real-time, ML-based recommendations. Option D is wrong because Cloud Dataflow is a stream processing service for data pipelines, not a pre-built recommendation engine; it can process clickstream data but cannot generate recommendations without additional custom ML model deployment.

397
Multi-Selectmedium

A company wants to run a critical stateful application on Compute Engine with the highest availability. The application requires block storage that can survive a zone failure. Which TWO actions should they take? (Choose TWO)

Select 2 answers
A.Use regional Persistent Disk
B.Place instances in a zonal managed instance group
C.Enable object versioning on a Cloud Storage bucket
D.Place instances in a regional managed instance group
E.Use zonal Persistent Disk
AnswersA, D

Regional Persistent Disk synchronously replicates data across two zones in the selected region, providing zone-failure resilience with write-consistent, read-after-write semantics. This makes it the correct choice for a critical stateful application because VM instances can be recreated or failed over to another zone while retaining the same disk data. It supports live migration, snapshots, and resizing, and is the recommended block storage for applications that require high availability without sacrificing durability.

Why this answer

Regional Persistent Disk replicates data across zones synchronously, surviving a zone failure. The instances should be in a regional managed instance group to distribute across zones. Zonal PD and single-zone MIG would not survive zone failure.

398
MCQhard

A traditional insurance company is facing competition from 'insurtech' startups that use telematics data, AI, and cloud platforms to offer usage-based, real-time personalized insurance products. The traditional company's CTO proposes a cloud-first digital transformation. Which business model change most clearly represents digital transformation rather than digitization?

A.Converting paper policy documents to digital PDFs stored in cloud document management systems
B.Replacing the claims processing fax machine with an online portal
C.Offering usage-based, dynamically priced insurance products using real-time telematics data and ML-driven individual risk assessment — replacing demographic-table pricing with behavioral data
D.Moving the company's email system to Google Workspace to improve employee collaboration
AnswerC

This is true transformation. The insurance product itself changes: pricing is no longer based on demographic averages but individual behavior. Good drivers pay less; high-risk behavior triggers real-time pricing changes. This requires cloud-scale real-time data processing and ML, and creates a fundamentally different customer value proposition from traditional insurance.

Why this answer

Option C is correct because digital transformation means fundamentally reimagining the business model and value proposition using digital technology — here, shifting from demographic-table pricing to real-time behavioral telematics data with ML-driven individual risk assessment creates an entirely new product category (usage-based insurance). This changes how value is created, priced, and delivered, not just how existing processes are executed. The other options merely digitize existing artifacts or channels.

Exam trap

GCDL often tests the distinction between digitization (converting analog to digital), digitalization (improving processes), and digital transformation (reinventing the business model) — candidates frequently pick process improvements like portals or PDFs thinking any digital initiative counts as transformation.

How to eliminate wrong answers

Option A is wrong because converting paper documents to PDFs is pure digitization — it changes the storage medium but not the business model, pricing, or customer value proposition. Option B is wrong because replacing a fax machine with an online portal is process digitization (channel modernization), not transformation; the claims process itself remains fundamentally the same. Option D is wrong because migrating email to Google Workspace is an IT infrastructure/productivity upgrade that improves internal collaboration but does not alter the insurance business model, underwriting logic, or customer offering.

399
MCQeasy

A startup wants to launch a new web application globally with minimal upfront investment and the ability to scale automatically based on traffic. Which cloud deployment model best meets these requirements?

A.On-premises private cloud
B.Public cloud
C.Hybrid cloud
D.Multi-cloud
AnswerB

Public cloud delivers compute, storage, and networking as metered, on-demand services from a provider whose infrastructure spans multiple regions and edge locations worldwide. A startup can provision resources in seconds, scale automatically in response to traffic, and only pay for what it consumes, avoiding upfront hardware costs. Global launch is simplified via provider-managed services like CDNs and load balancers, and the operational model shifts to pay-as-you-go, ideal for uncertain demand.

Why this answer

Public cloud offers on-demand resources, pay-as-you-go pricing, and global reach without upfront capital expenditure.

400
MCQmedium

A retail company runs its order-processing application on Compute Engine instances in a single zone. During a regional maintenance event, the zone becomes unavailable for several hours, and the application goes down. The operations team wants to improve resilience so that a single zone failure does not cause an outage, while keeping changes minimal. What should they do?

A.Enable live migration for all instances so that they automatically move to another zone when the current zone fails.
B.Configure a Cloud VPN tunnel between the affected zone and a second zone, and replicate the application state over that tunnel.
C.Move the application to a single, larger instance with a higher availability SLA so that it is less likely to fail.
D.Create a managed instance group with instances spread across multiple zones in the same region, and configure an HTTP(S) load balancer to distribute traffic.
AnswerD

Spreading instances across multiple zones in a managed instance group provides redundancy against a single-zone failure. An HTTP(S) load balancer with a global anycast IP routes users to healthy instances, so if one zone fails, traffic continues to others. This directly addresses the requirement to survive a zone outage with minimal architectural change.

Why this answer

The most effective way to survive a single-zone failure is to distribute instances across multiple zones within a region using a managed instance group, and to use a load balancer to route traffic to healthy instances. This design removes the single point of failure and aligns with Google Cloud's regional resource model. Other options either do not address zone-level failures or add unnecessary components.

Exam trap

The trap here is assuming that live migration protects against zone failures, when it only moves VMs between hosts within the same zone.

401
MCQeasy

A product manager wants to understand what 'latency' means for her company's cloud-hosted e-commerce application. Her developer explains that latency is critical for user experience. Which definition of latency is most accurate in this context?

A.Latency is the total amount of data that can be transferred per second between the user and the application
B.Latency is the time elapsed between a user action (click, page load) and receiving the server's response — directly affecting how fast and responsive the application feels
C.Latency is the percentage of time the application is available versus unavailable
D.Latency is the number of requests the server can handle simultaneously before performance degrades
AnswerB

This correctly defines latency in the context of web applications. High latency makes applications feel slow and unresponsive. For e-commerce, high latency directly increases cart abandonment. Techniques like CDN, edge computing, and database query optimization reduce latency.

Why this answer

Latency in the context of a cloud-hosted e-commerce application specifically measures the round-trip time from a user action (such as a click or page load) to the receipt of the server's response. This directly impacts perceived responsiveness and user experience, as higher latency leads to noticeable delays in interactions like adding items to a cart or checking out.

Exam trap

Google Cloud often tests the distinction between latency and throughput, trapping candidates who confuse the total data transfer rate (bandwidth) with the time delay of a single transaction.

How to eliminate wrong answers

Option A is wrong because it describes throughput (bandwidth), not latency; throughput measures data transfer rate per second, while latency measures delay. Option C is wrong because it defines availability (uptime), often expressed as a percentage of time the service is operational, not latency. Option D is wrong because it describes concurrency or capacity (the number of simultaneous requests a server can handle), which is related to scalability and load handling, not the time delay of a single request-response cycle.

402
MCQmedium

A company wants to use Google Cloud for a workload that requires compliance with PCI DSS. Which of the following is a Google responsibility under the shared model?

A.Configuring firewall rules to protect cardholder data
B.Managing access keys for cardholder data
C.Encrypting cardholder data at rest
D.Patching the hypervisor
AnswerD

Patching the hypervisor is a Google responsibility because the hypervisor is a core component of the infrastructure that hosts all customer virtual machines and is fully managed by Google. Google performs security patches and updates to the hypervisor without customer involvement, ensuring that vulnerabilities in the virtualization layer are addressed. This is a clear example of where the cloud provider, not the customer, maintains the security of the foundational compute environment.

Why this answer

Google is responsible for maintaining the security of the infrastructure, including the hypervisor, network, and physical security, which are part of PCI DSS compliance scope for the cloud provider.

403
Multi-Selectmedium

Which TWO Google Cloud services help prevent data exfiltration from virtual machines?

Select 2 answers
A.Access Transparency
B.Security Command Center
C.Cloud Armor
D.Cloud Data Loss Prevention (DLP)
E.VPC Service Controls
AnswersD, E

Cloud Data Loss Prevention (DLP) uses content inspection with built-in detectors to identify sensitive elements like credit card numbers, PII, or credentials, and can apply transformations such as redaction, masking, tokenization, and encryption. When integrated into data pipelines or egress workflows, it can block or de-identify sensitive content before it leaves your organization, directly preventing data exfiltration at the data layer.

Why this answer

Cloud Data Loss Prevention (DLP) is correct because it enables inspection of data at rest and in motion for sensitive content (e.g., PII, credit card numbers) using predefined or custom infoTypes. When integrated with VPC Service Controls, it can block or redact sensitive data before it leaves the virtual machine's network boundary, directly preventing data exfiltration.

Exam trap

Google Cloud often tests the distinction between monitoring/logging services (like Access Transparency and Security Command Center) and active data exfiltration prevention controls (like VPC Service Controls and Cloud DLP), leading candidates to select options that only provide visibility rather than enforcement.

404
MCQmedium

A company has migrated sensitive customer data to Google Cloud. The legal team asks: 'If Google is hosting our data, who is responsible for ensuring that data is not improperly accessed by unauthorized users through our application?' Under the shared responsibility model, how should the CTO answer?

A.Google is fully responsible because they host the data and control the infrastructure
B.The customer is responsible for application access controls, authentication, and IAM policies that protect data from unauthorized application-layer access, while Google secures the underlying infrastructure
C.Both Google and the customer share equal 50/50 responsibility for all data access controls
D.No one is responsible because cloud computing inherently cannot prevent unauthorized access
AnswerB

This is the correct shared responsibility answer. Google secures the infrastructure layer — physical hardware, network, hypervisor. The customer must secure their application layer: who can access the application, how they authenticate, what permissions their service accounts have, and whether the application has vulnerabilities.

Why this answer

Under the Google Cloud shared responsibility model, the customer is responsible for securing access to their application and data, including authentication, authorization, and IAM policies, while Google is responsible for the security of the underlying infrastructure (physical security, network, hypervisor). The legal team's question specifically asks about unauthorized access through the customer's application, which falls under the customer's responsibility for application-layer controls.

Exam trap

The trap here is that candidates often assume the cloud provider is fully responsible for all security aspects, but the shared responsibility model explicitly places application-layer access controls, authentication, and IAM on the customer, especially when the question specifies 'through our application'.

How to eliminate wrong answers

Option A is wrong because it incorrectly states Google is fully responsible for all data access; Google secures the infrastructure but the customer must manage application-layer access controls, IAM, and authentication. Option C is wrong because responsibility is not a fixed 50/50 split; it is a shared model where Google secures the infrastructure and the customer secures their data, applications, and access policies. Option D is wrong because cloud computing can prevent unauthorized access through proper implementation of security controls like IAM, encryption, and network policies; it is not inherently incapable of preventing unauthorized access.

405
MCQhard

A company wants to use Google Cloud's Generative AI capabilities to build an internal assistant that can answer questions about company policies using documents stored in Google Drive. Which Google Cloud product provides pre-built infrastructure for building this type of AI application?

A.BigQuery ML — train a custom language model on company policy documents.
B.Vertex AI Agent Builder with Gemini and document-grounded search (RAG).
C.Cloud Natural Language API — it reads and summarizes documents automatically.
D.Cloud Translation API — it translates policy documents into the user's language.
AnswerB

Vertex AI Agent Builder is correct because it provides pre-built RAG (Retrieval-Augmented Generation) pipelines: you can ingest policy documents from sources like Google Drive or Cloud Storage, index them for semantic retrieval, and ground Gemini's responses in that private document content. This eliminates the need for ML expertise or custom model training while ensuring answers are factually tied to your policies, reducing hallucination. The service handles document chunking, vector search, and LLM orchestration out of the box, making it the ideal low-code solution for building a document Q&A assistant.

Why this answer

Vertex AI Agent Builder with Gemini and document-grounded search (RAG) is correct because it provides pre-built infrastructure for building a generative AI assistant that retrieves information from enterprise documents. It combines Gemini's large language model with Retrieval-Augmented Generation (RAG) to ground answers in company policy documents stored in Google Drive, without requiring custom model training or manual infrastructure setup.

Exam trap

The GCDL exam often tests the distinction between pre-built AI application infrastructure (Vertex AI Agent Builder) and individual AI/ML services (like BigQuery ML, Natural Language API, or Translation API) that require custom integration to build a complete assistant.

How to eliminate wrong answers

Option A is wrong because BigQuery ML is designed for training custom machine learning models using SQL queries on structured data in BigQuery, not for building a pre-built generative AI assistant with document retrieval from Google Drive. Option C is wrong because Cloud Natural Language API provides pre-trained models for entity extraction, sentiment analysis, and syntax analysis, but it does not offer a pre-built infrastructure for building a conversational AI assistant with RAG-based document grounding. Option D is wrong because Cloud Translation API is a service for translating text between languages, not for building a question-answering assistant that retrieves and reasons over company policy documents.

406
MCQhard

A company has a requirement from their security auditor to demonstrate that all administrative actions performed in Google Cloud (such as creating VMs, modifying IAM policies, and deleting storage buckets) are logged and tamper-evident. Which Cloud Logging log type fulfills this requirement?

A.Data Access audit logs — they capture all read and write operations.
B.Admin Activity audit logs — always-on, tamper-resistant logs of all administrative API calls.
C.System Event audit logs — they capture all Google Cloud operations.
D.Cloud Monitoring logs — they track all changes to monitored resources.
AnswerB

Admin Activity audit logs are automatically enabled for all Google Cloud projects and cannot be disabled or deleted by any user, including project owners or Org admins. They record all control-plane API calls such as resource creation, modification, and deletion, along with the calling principal, the request parameters, and the response status. Because these logs are immutable and stored independently of user access, they provide a tamper-resistant, authoritative record that satisfies audit requirements.

Why this answer

Admin Activity audit logs are always-on, tamper-resistant logs that record all administrative API calls, such as creating VMs, modifying IAM policies, and deleting storage buckets. They cannot be disabled or modified by users, ensuring tamper-evident logging for security auditor requirements.

Exam trap

The trap here is that candidates confuse Data Access audit logs (which require enabling and capture data-level operations) with Admin Activity audit logs (which are always-on and capture administrative actions), leading them to incorrectly select option A.

How to eliminate wrong answers

Option A is wrong because Data Access audit logs capture read and write operations on user data, not administrative actions like creating VMs or modifying IAM policies, and they are not always-on (they must be explicitly enabled). Option C is wrong because System Event audit logs capture Google Cloud system events (e.g., automatic maintenance), not administrative actions performed by users. Option D is wrong because Cloud Monitoring logs track metrics and alerts for resource performance, not administrative API calls, and they are not tamper-resistant logs.

407
MCQmedium

A company is migrating to Google Cloud and wants to reduce operational overhead for managing their infrastructure. Which Google Cloud service allows them to define infrastructure as code and automate provisioning?

A.Cloud Deployment Manager
B.Google Cloud SDK
C.Cloud Console
D.Cloud Shell
AnswerA

Cloud Deployment Manager is Google Cloud's native Infrastructure as Code (IaC) service, allowing you to define your entire infrastructure in declarative YAML or Python templates. Once deployed, it treats templates as the source of truth, handling incremental updates, dependencies, and idempotent rollbacks automatically. This aligns perfectly with a migration goal to reduce operational overhead by making resource provisioning repeatable, auditable, and version-controllable.

Why this answer

Cloud Deployment Manager is the correct answer because it is a Google Cloud service that allows you to define your infrastructure as code using declarative templates (in YAML, Python, or Jinja2). It automates the provisioning and management of Google Cloud resources, reducing manual operational overhead by enabling repeatable, version-controlled deployments.

Exam trap

The trap here is that candidates confuse Cloud Deployment Manager with general-purpose tools like Cloud SDK or Cloud Shell, assuming any command-line or scripting tool can achieve infrastructure-as-code automation, but only Deployment Manager provides declarative, managed provisioning.

How to eliminate wrong answers

Option B (Google Cloud SDK) is wrong because it is a command-line toolset for interacting with Google Cloud services, not a service for defining infrastructure as code or automating provisioning. Option C (Cloud Console) is wrong because it is a web-based GUI for managing resources manually, which does not support infrastructure-as-code definitions or automated provisioning. Option D (Cloud Shell) is wrong because it is a browser-based terminal environment with pre-installed tools, not a service for defining or automating infrastructure deployment.

408
Multi-Selecteasy

A developer wants to deploy a serverless application that runs code in response to HTTP requests and events from other Google Cloud services. They also need to store configuration and session data in a fast, in-memory data store. Which TWO services should they use? (Choose TWO)

Select 2 answers
A.Cloud Functions
B.Cloud SQL
C.Memorystore
D.Cloud Bigtable
E.Cloud Storage
AnswersA, C

Cloud Functions is Google Cloud's serverless compute service, executing code in response to HTTP requests and events from other Google Cloud services without server management. It satisfies the stem's event-driven, serverless requirement directly, pairing with Memorystore for in-memory session data.

Why this answer

Cloud Functions handles HTTP and event-driven triggers serverlessly. Memorystore provides managed Redis/Memcached for caching and session storage. Cloud SQL is relational and not in-memory.

Cloud Storage is object storage. Bigtable is NoSQL but not in-memory.

409
MCQeasy

A company has a Google Cloud environment with 50 projects and 200 engineers. The security team wants to ensure that a new security policy — requiring all Cloud Storage buckets to have uniform bucket-level access enabled — applies to all existing and future buckets across all projects. Which approach scales to the entire organization?

A.Send an email to all 200 engineers explaining the policy and asking them to manually enable uniform bucket-level access on their buckets
B.Apply an Organization Policy constraint ('storage.uniformBucketLevelAccess') at the organization level to enforce the setting automatically across all current and future projects and buckets
C.Create a Cloud Function that checks bucket configurations hourly and enables uniform access on non-compliant buckets
D.Grant the security team Owner access to all 50 projects so they can manually enforce the policy in each project
AnswerB

Organization Policy is the scalable solution. By applying the constraint at the organization level, it cascades to all 50 projects automatically. New projects created in the future also inherit the constraint. No per-project configuration or per-engineer action required.

Why this answer

Organization Policy constraints, such as `storage.uniformBucketLevelAccess`, are enforced at the organization level and automatically apply to all existing and future projects and resources within the organization. This ensures uniform compliance without manual intervention, scaling seamlessly across 50 projects and 200 engineers.

Exam trap

The GCDL exam often tests the distinction between reactive remediation (e.g., Cloud Functions) and proactive enforcement (e.g., Organization Policies), where candidates may choose a technically functional but less scalable or secure option like C because it seems automated, missing the requirement for organization-wide, preventive enforcement.

How to eliminate wrong answers

Option A is wrong because relying on manual action from 200 engineers is error-prone, unscalable, and does not guarantee enforcement for future buckets. Option C is wrong because a Cloud Function that periodically checks and remediates buckets is reactive, not preventive, and introduces latency and potential gaps between checks; it also does not enforce the policy on new buckets before they are created. Option D is wrong because granting Owner access to the security team for all 50 projects violates the principle of least privilege, creates a security risk, and still requires manual effort to apply the policy to each bucket, which does not scale.

410
MCQmedium

A company wants to create a customer-facing conversational AI assistant that understands natural language and can answer questions about its products, integrated into their website and mobile app. Which Google Cloud AI product is the most appropriate starting point?

A.BigQuery ML, for building machine learning models on customer data to predict product recommendations
B.Dialogflow CX, Google Cloud's managed conversational AI platform for building natural language understanding chatbots integrated into web and mobile apps
C.Cloud Vision API, for analyzing images to understand customer product photos
D.Cloud Natural Language API, for analyzing the sentiment of customer product reviews
AnswerB

Dialogflow CX is purpose-built for conversational AI: it handles NLU (natural language understanding), dialog flow management, intent detection, entity extraction, and integrations with multiple channels (web widget, mobile, messaging apps). It requires no ML expertise to build effective conversational agents.

Why this answer

Dialogflow CX is the correct choice because it is Google Cloud's managed conversational AI platform specifically designed for building natural language understanding (NLU) chatbots that can be integrated into websites and mobile apps. It provides advanced state management, flow-based conversation design, and seamless integration with web and mobile channels, making it the most appropriate starting point for a customer-facing conversational AI assistant.

Exam trap

Google Cloud often tests the distinction between a full conversational AI platform (Dialogflow CX) and individual AI APIs (like Cloud Natural Language API or Cloud Vision API) that perform only a single task, leading candidates to mistakenly choose a component API instead of the integrated platform.

How to eliminate wrong answers

Option A is wrong because BigQuery ML is a tool for building and deploying machine learning models using SQL on data stored in BigQuery, not for building conversational AI or natural language understanding chatbots. Option C is wrong because Cloud Vision API is an image analysis service that extracts information from images, such as labels, text, and faces, and has no capability for natural language understanding or conversational interactions. Option D is wrong because Cloud Natural Language API provides pre-trained models for sentiment analysis, entity recognition, and syntax analysis, but it is not a conversational AI platform and cannot manage multi-turn dialogues, state, or integrate directly as a chatbot.

411
MCQhard

A regional insurance company competes with an InsurTech startup that uses cloud-native AI to personalize policies, process claims in minutes, and launch new products weekly. The traditional insurer takes 6 months to launch new products and 2 weeks to process claims. Which cloud-enabled business model advantage does the startup have?

A.Lower insurance premiums because cloud infrastructure costs less than data centers.
B.Innovation velocity and operational efficiency through cloud-native AI, enabling faster product iteration and dramatically faster customer service delivery.
C.Better regulatory compliance because cloud providers have more compliance certifications.
D.Access to more insurance actuarial data than the traditional insurer.
AnswerB

Cloud-native AI services such as serverless ML inference, pre-trained APIs, and auto-scaled compute allow the startup to embed intelligent decision-making directly into its customer service workflow, compressing claims processing from two weeks to minutes. Weekly product releases, versus the traditional insurer's six-month cycles, become feasible because the cloud removes upfront infrastructure procurement and supports continuous deployment pipelines. This combination of innovation velocity and AI-driven operational efficiency is what creates a sustainable competitive advantage, not merely having access to cloud resources.

Why this answer

The startup leverages cloud-native AI to achieve innovation velocity (weekly product launches vs. 6 months) and operational efficiency (minutes vs. 2 weeks for claims). This is a direct cloud-enabled business model advantage: elastic infrastructure and AI services allow rapid iteration and automated workflows, which traditional on-premises systems cannot match.

Exam trap

The GCDL exam often tests the misconception that cloud adoption is primarily about cost savings (Option A) rather than business agility and innovation velocity, which are the true transformative advantages in this scenario.

How to eliminate wrong answers

Option A is wrong because cloud infrastructure does not inherently lower premiums; cost savings depend on usage optimization and are not guaranteed, and the question focuses on speed and agility, not cost. Option C is wrong because while cloud providers offer compliance certifications, regulatory compliance is not a unique advantage—traditional insurers can also achieve compliance, and the startup's edge is speed, not compliance. Option D is wrong because access to actuarial data is not a cloud-native advantage; data access depends on partnerships and data sources, not the cloud platform itself.

412
MCQmedium

A developer wants to deploy a Python script that runs in response to new files uploaded to a Cloud Storage bucket. The script performs simple image transformations. Which compute service is the BEST fit?

A.App Engine
B.Cloud Functions
C.Compute Engine
D.Cloud Run
AnswerB

Cloud Functions is Google Cloud's event-driven serverless compute platform that runs Python code in response to specific triggers, including Cloud Storage events like object finalization. With a Storage trigger, the function is automatically invoked with the object metadata, eliminating the need to manage infrastructure or a web server. This matches the requirement to react to file uploads with low operational overhead and automatic scaling.

Why this answer

Cloud Functions is event-driven and designed for lightweight code that runs in response to events like Cloud Storage object changes.

413
MCQhard

An organization's security team reviews their Google Cloud environment and finds that several Cloud Storage buckets have `allAuthenticatedUsers` bindings, and multiple service accounts have the Owner role. Which Google Cloud tool automatically identifies these types of high-risk IAM configurations?

A.Cloud Audit Logs — reviewing all recent IAM changes.
B.Security Command Center (SCC) with IAM Recommender and Security Health Analytics.
C.Cloud Billing reports — they flag expensive configurations that indicate security issues.
D.Cloud Monitoring — it alerts when IAM policies are modified.
AnswerB

Security Command Center (SCC) with Security Health Analytics and IAM Recommender is the correct choice because it provides continuous, automated scanning for high-risk IAM misconfigurations such as public Cloud Storage buckets, overly broad roles like Owner or Editor, and over-permissive service accounts. Security Health Analytics uses built-in detectors to evaluate the current state of your GCP resources, while IAM Recommender analyzes actual usage patterns to generate actionable recommendations for reducing permissions to least privilege. This combination proactively surfaces existing weaknesses and delivers concrete remediation steps, making it a comprehensive and proactive security posture management tool.

Why this answer

Security Command Center (SCC) with Security Health Analytics and IAM Recommender is the correct tool because it automatically scans for high-risk IAM configurations, such as `allAuthenticatedUsers` bindings on Cloud Storage buckets and service accounts with the Owner role. Security Health Analytics detects misconfigurations against CIS benchmarks and Google Cloud best practices, while IAM Recommender provides actionable recommendations to reduce excessive permissions. This combination proactively identifies and helps remediate these specific security risks without requiring manual log review or billing analysis.

Exam trap

The trap here is that candidates often confuse Cloud Audit Logs or Cloud Monitoring with proactive security scanning tools, not realizing that those services only provide raw data or alerts on changes, whereas SCC with Security Health Analytics and IAM Recommender actively analyzes the configuration state to detect high-risk IAM bindings.

How to eliminate wrong answers

Option A is wrong because Cloud Audit Logs only record historical IAM changes and require manual review to identify high-risk configurations; they do not automatically detect or flag misconfigurations like `allAuthenticatedUsers` bindings. Option C is wrong because Cloud Billing reports focus on cost analysis and do not have the capability to identify IAM security misconfigurations or excessive permissions. Option D is wrong because Cloud Monitoring can alert on policy changes via logs-based metrics, but it does not natively analyze the content of IAM policies to detect high-risk bindings like `allAuthenticatedUsers` or Owner roles on service accounts.

414
MCQmedium

A multinational company must ensure that personal data of European citizens stored in Google Cloud cannot be accessed by or transferred to systems outside the European Union, as required by GDPR data residency requirements. Which Google Cloud controls most directly enforce this?

A.Enabling HTTPS for all data transmission to ensure data is encrypted during transfer
B.Configuring organization policy to restrict resource creation to EU regions, using VPC Service Controls to prevent data movement outside the EU perimeter, and establishing a GDPR-compliant Data Processing Agreement with Google
C.Using Customer-Managed Encryption Keys (CMEK) where the encryption keys are stored outside Google's infrastructure
D.Training developers about GDPR requirements and requiring manual approval for any cross-region data transfers
AnswerB

This combination addresses GDPR data residency: org policy constraints prevent resources from being created outside EU regions; VPC Service Controls prevent data from being read out of the EU perimeter; the DPA provides contractual compliance assurance. Together they form a comprehensive GDPR data residency control framework.

Why this answer

It combines three essential controls that directly enforce GDPR data residency: Organization Policies restrict resource creation to EU regions, VPC Service Controls create a data perimeter preventing exfiltration outside the EU, and a GDPR-compliant Data Processing Agreement (DPA) establishes the legal framework for data handling. These controls work together to ensure data at rest and in transit remains within the EU boundary, directly addressing the residency requirement.

Exam trap

Google Cloud often tests the misconception that encryption (HTTPS or CMEK) alone satisfies data residency requirements, when in fact residency is about geographic location of data, not its confidentiality during transit or at rest.

How to eliminate wrong answers

Option A is wrong because HTTPS only encrypts data in transit; it does not restrict where data is stored or prevent it from being transferred to systems outside the EU, so it fails to enforce data residency. Option C is wrong because CMEK controls encryption key management, not data location; keys stored outside Google's infrastructure do not prevent data from being moved or stored outside the EU, and GDPR residency is about physical location, not key custody. Option D is wrong because training and manual approval are procedural controls, not technical enforcement mechanisms; they rely on human compliance and cannot guarantee that data never leaves the EU, especially at scale or in automated environments.

415
MCQmedium

A company uses committed use discounts (CUDs) for its production workload baseline. An engineer proposes also using sustained use discounts (SUDs) for the same VMs. Why is this incorrect?

A.CUDs and SUDs can be combined on the same VMs — applying both gives the maximum possible discount
B.CUDs and SUDs are mutually exclusive: VMs already covered by committed use discounts don't accrue sustained use discounts — you receive only the CUD, not both
C.SUDs cannot be applied to production workloads — they are only available for development environments
D.Applying both CUDs and SUDs creates a billing conflict that could result in Google charging the company more than on-demand pricing
AnswerB

This is correct. CUDs are pre-purchased commitments that replace (not supplement) the SUD credit system. When a CUD commitment covers compute usage, that usage is billed at the CUD rate, not the on-demand rate that would otherwise accumulate SUD credits. Stacking is not possible.

Why this answer

Committed use discounts (CUDs) and sustained use discounts (SUDs) are mutually exclusive on the same VM. When a VM is covered by a CUD, it does not accrue SUDs; only the CUD discount is applied. This prevents double-discounting and ensures billing consistency.

Exam trap

The trap here is that candidates may assume discounts are additive or combinable, similar to how some cloud providers allow stacking, but Google Cloud explicitly makes CUDs and SUDs mutually exclusive to prevent double-discounting.

How to eliminate wrong answers

Option A is wrong because CUDs and SUDs cannot be combined on the same VMs; they are mutually exclusive, so applying both does not give the maximum possible discount. Option C is wrong because SUDs are available for all workloads, including production, not just development environments. Option D is wrong because applying both CUDs and SUDs does not create a billing conflict that results in higher charges than on-demand pricing; instead, the system simply applies only the CUD and ignores SUD accrual.

416
MCQmedium

An operations team is performing a post-incident review after a production outage. The team lead insists that the review must follow a 'blameless postmortem' approach. What does this mean, and why is it important for organizational learning?

A.A blameless postmortem assigns full responsibility to the automated systems involved, not to human engineers, which protects the team from accountability
B.A blameless postmortem focuses on systemic root causes and improvement opportunities rather than individual fault — creating psychological safety for honest disclosure and leading to more effective prevention of future incidents
C.A blameless postmortem means the incident is not formally documented to protect employees' privacy and career records
D.A blameless postmortem can only be conducted by senior management who have authority to make systemic improvements
AnswerB

This captures both dimensions: what blameless means (systemic focus, not individual blame) and why it matters (psychological safety enables honest disclosure — people share full details when they don't fear punishment). SRE culture pioneered this approach, which produces better learning than punitive reviews.

Why this answer

A blameless postmortem in Google Cloud operations (and SRE practice) shifts focus from individual human error to systemic root causes, such as misconfigured alerting thresholds, insufficient canary deployments, or gaps in monitoring coverage. This approach fosters psychological safety, encouraging engineers to report all contributing factors without fear of reprisal, which leads to more effective incident prevention and aligns with Google's Site Reliability Engineering (SRE) principles of learning from failures.

Exam trap

Google Cloud often tests the misconception that 'blameless' means 'no accountability' or 'no documentation', but the correct understanding is that it shifts accountability from individuals to systemic improvements while still requiring thorough documentation and follow-up actions.

How to eliminate wrong answers

Option A is wrong because a blameless postmortem does not assign responsibility to automated systems; instead, it examines both human and system factors to identify systemic improvements, and it does not protect the team from accountability—it promotes accountability for learning. Option C is wrong because a blameless postmortem is formally documented (e.g., in a postmortem template stored in Google Cloud Storage or a shared drive) to capture findings and action items, not to protect privacy or career records—privacy is a side effect, not the purpose. Option D is wrong because a blameless postmortem can be conducted by any team member, including individual contributors, not only senior management; the goal is to involve those closest to the incident for accurate root cause analysis.

417
MCQmedium

A company wants to analyse streaming data from IoT devices in real time with sub-second latency, using SQL queries. Which combination of services should they use?

A.Cloud IoT Core + Cloud Functions + Bigtable
B.Cloud Pub/Sub + Dataproc + Cloud Storage
C.Cloud Pub/Sub + Cloud Functions + Cloud SQL
D.Cloud Pub/Sub + Dataflow + BigQuery
AnswerD

Pub/Sub ingests the IoT data stream durably and asynchronously, Dataflow (Apache Beam) processes it in a fully managed, autoscaling manner with sub-second latency, exactly-once semantics, and support for event-time windows, filters, and enrichments, then writes results to BigQuery using the Storage Write API. BigQuery is a fully managed, serverless, columnar data warehouse that provides native SQL analytics over the streamed data, making this combination the canonical GCP architecture for real-time streaming SQL.

Why this answer

Dataflow with unbounded sources (like Pub/Sub) and SQL via Beam SQL or Dataflow SQL can process streaming data with low latency. BigQuery can also stream data but with higher latency (seconds). Cloud Functions is not ideal for real-time SQL analytics.

Dataproc is for batch processing.

418
MCQmedium

An organization wants to ensure business continuity by failing over to a secondary region in case of disaster. Which cloud characteristic enables this capability?

A.Global infrastructure
B.Scalability
C.Measured service
D.Resource pooling
AnswerA

Global infrastructure is the correct answer because Google Cloud's global footprint spans multiple independent regions, each with at least three zones. For business continuity, these regions enable cross-region failover: if an entire region experiences an outage, workloads and data can be redirected to a healthy region. This geographic redundancy is designed precisely to support disaster recovery scenarios, unlike the other listed characteristics.

Why this answer

Global infrastructure refers to the distributed network of data centers across multiple geographic regions. By deploying resources in a secondary region, an organization can fail over to that region during a disaster, ensuring business continuity. This capability is unique to the cloud's global footprint, not to other characteristics like scalability or measured service.

Exam trap

Google Cloud often tests the misconception that scalability or resource pooling alone can provide disaster recovery, but failover requires the physical presence of infrastructure in separate geographic regions, which is a function of global infrastructure.

How to eliminate wrong answers

Option B is wrong because scalability refers to the ability to increase or decrease resources on demand, not to geographic redundancy for disaster recovery. Option C is wrong because measured service involves metering resource usage for billing and optimization, not failover capabilities. Option D is wrong because resource pooling allows multiple customers to share physical resources via virtualization, but does not inherently provide cross-region failover or disaster recovery.

419
MCQmedium

A team needs to run a machine learning model using custom code in Python with TensorFlow, and they want to train it at scale on GPU hardware without managing infrastructure. Which Google Cloud service is best suited?

A.Vertex AI
B.Cloud Run
C.Compute Engine
D.Cloud Functions
AnswerA

Vertex AI is a fully managed machine learning platform that supports custom container training, allowing you to bring your own model code and dependencies while leveraging managed GPU and TPU clusters. It provides automatic scaling, hyperparameter tuning, and integrated MLOps tools like model versioning, monitoring, and Vertex Pipelines, which drastically reduce operational overhead compared to raw compute. This makes it the ideal choice for running custom ML models at scale without manually provisioning infrastructure.

Why this answer

Vertex AI is Google Cloud's unified ML platform that supports custom Python code with frameworks like TensorFlow and provides managed training on GPU hardware without infrastructure management. It offers Vertex AI Training with pre-built containers or custom containers, auto-scaling, and integration with GPUs/TPUs. This directly matches the requirement to train at scale on GPUs with custom code.

Exam trap

GCDL often tests the distinction between managed ML platforms and raw compute, and candidates pick Compute Engine because it 'supports GPUs'—missing that the requirement for no infrastructure management points to Vertex AI.

How to eliminate wrong answers

Option B is wrong because Cloud Run is a serverless container platform for stateless HTTP services, not designed for GPU-accelerated ML training jobs or long-running training workloads. Option C is wrong because Compute Engine provides raw VMs where the team would have to manage infrastructure, install drivers, and orchestrate training—contradicting the 'without managing infrastructure' requirement. Option D is wrong because Cloud Functions is an event-driven serverless compute service for short-lived functions, with no GPU support and execution time limits that make it unsuitable for ML training.

420
MCQeasy

An organization is considering cloud adoption. Their CTO argues that 'the cloud is just someone else's computers — why should we trust it?' Which is the strongest counterargument for cloud trust and reliability?

A.Cloud providers can be fully trusted because governments require them to guarantee zero downtime.
B.Google Cloud operates at a scale enabling reliability (multiple 9s SLAs, redundant infrastructure, third-party audits) that most organizations cannot achieve with their own data centers.
C.Google employees are more trustworthy than the company's own IT staff.
D.The CTO's concern is valid — companies should never move sensitive data to the cloud.
AnswerB

Google Cloud's global infrastructure spans multiple regions and zones, with redundant power, networking, storage, and cooling, plus a private fiber backbone and dedicated reliability engineering teams. This scale yields industry-leading SLAs (multiple 9s), and independent third-party audits (ISO 27001, SOC 2) verify the effectiveness of security and resilience controls. Most organizations would find it economically and operationally infeasible to replicate this level of redundancy, expertise, and compliance testing in their own data centers.

Why this answer

It directly addresses the CTO's concern by highlighting that major cloud providers like Google Cloud operate at a scale that enables reliability metrics (e.g., 99.99% uptime SLAs) and infrastructure redundancy (e.g., multi-region deployments, automatic failover) that most on-premises data centers cannot match. This is supported by third-party audits (e.g., SOC 2, ISO 27001) that validate security and operational practices, making the cloud not just 'someone else's computers' but a professionally managed, highly resilient environment.

Exam trap

The trap here is that candidates may choose Option A because they overestimate government mandates or SLAs as guarantees of zero downtime, or Option D because they confuse valid caution with outright rejection, missing the nuanced argument that cloud providers offer superior reliability through scale and professional management.

How to eliminate wrong answers

Option A is wrong because governments do not require cloud providers to guarantee zero downtime; SLAs typically offer service credits for downtime but never guarantee 100% uptime, and zero downtime is technically impossible due to factors like planned maintenance or unforeseen outages. Option C is wrong because it makes an unfounded generalization about trustworthiness; cloud providers rely on strict access controls, encryption, and compliance frameworks (e.g., IAM, data encryption at rest and in transit) rather than personal trust, and employees of any organization can be vetted similarly. Option D is wrong because it dismisses cloud adoption entirely without considering risk mitigation strategies like data encryption, access management, and compliance certifications that make sensitive data secure in the cloud; many regulated industries (e.g., healthcare, finance) successfully use cloud services with proper controls.

421
MCQhard

A global gaming company needs to serve low-latency content to players worldwide. They want to cache static assets (images, videos) at edge locations closest to users. Which Google Cloud product should they use?

A.Cloud Storage
B.Cloud CDN
C.Compute Engine with global load balancer
D.Cloud Spanner
AnswerB

Cloud CDN caches static assets at Google's globally distributed edge points of presence, serving images and video from locations nearest each player. This reduces origin round trips and latency, meeting the worldwide low-latency caching requirement.

Why this answer

Cloud CDN uses Google's global edge network Points of Presence (PoPs) to cache content, reducing latency for users worldwide.

422
MCQmedium

A retail bank is building a partnership with a fintech startup. The bank provides regulated financial services infrastructure and customer reach; the fintech provides innovative digital experiences. Which cloud architectural pattern most naturally enables this kind of bank-fintech partnership?

A.Giving the fintech startup direct database access to the bank's customer records system for maximum data sharing
B.An API-first Open Banking architecture where the bank exposes regulated capabilities (accounts, payments, KYC) through managed APIs that the fintech builds innovative experiences on top of
C.The bank should acquire the fintech startup and consolidate all technology onto the bank's legacy infrastructure
D.The fintech should build all required banking infrastructure independently to avoid dependency on the bank's legacy systems
AnswerB

This is the Open Banking / BaaP pattern. The bank's APIs provide the regulated foundation (PSD2, open banking standards); the fintech builds customer-facing innovation on top. API management (like Apigee) provides authentication, rate limiting, versioning, and analytics for the partnership. This is exactly how modern bank-fintech partnerships work.

Why this answer

An API-first Open Banking architecture allows the bank to expose regulated capabilities (e.g., account information, payment initiation, KYC verification) through managed, secure APIs. The fintech can then build innovative digital experiences on top of these APIs without direct access to the bank's core systems, ensuring compliance, security, and loose coupling. This pattern aligns with PSD2 and Open Banking standards, enabling partnership without compromising regulatory control.

Exam trap

The trap here is that candidates may confuse 'data sharing' with 'direct database access' (Option A), failing to recognize that secure, API-mediated access is the correct architectural pattern for regulated partnerships, not raw data exposure.

How to eliminate wrong answers

Option A is wrong because giving the fintech direct database access to the bank's customer records system violates data privacy regulations (e.g., GDPR, PCI DSS) and creates severe security risks, as the fintech would have unfettered access to sensitive data without the bank's governance layer. Option C is wrong because acquiring the fintech and consolidating onto legacy infrastructure defeats the purpose of the partnership—it eliminates the fintech's agility and innovation, and legacy systems typically lack modern API capabilities, leading to technical debt and slower time-to-market. Option D is wrong because having the fintech build all required banking infrastructure independently is impractical and inefficient; it duplicates regulated capabilities (e.g., obtaining banking licenses, building secure transaction processing) that the bank already provides, negating the partnership's synergy and increasing cost and compliance burden.

423
MCQmedium

A traditional bank processes loan applications using manual paper-based workflows that take 2 weeks per application. The bank wants to use cloud technology to reduce this to under 24 hours. Which cloud-enabled capability primarily drives this transformation?

A.Lower storage costs for paper documents by digitizing them in Cloud Storage.
B.Cloud-based AI/ML services and workflow automation that process applications end-to-end without manual steps.
C.Moving the bank's email system to a cloud-based provider.
D.Using Cloud SQL instead of on-premises Oracle database.
AnswerB

Managed AI services such as Document AI extract key-value pairs from loan applications and financial statements, while auto-scaled workflow systems (e.g., Cloud Workflows or Dataflow) route verified data to risk-scoring models and trigger conditional approvals. This eliminates manual data entry, cross-department handoffs, and human review bottlenecks, shrinking processing time from weeks to hours. Because the entire end-to-end pipeline runs server-free, decisions can be made in near-real-time without human intervention.

Why this answer

Cloud-based AI/ML services combined with workflow automation can process loan applications end-to-end without manual intervention, reducing processing time from 2 weeks to under 24 hours. This transformation is driven by the ability to automate document extraction, validation, and decision-making using services like Google Cloud Document AI and Workflows, which eliminate the bottleneck of manual paper-based workflows.

Exam trap

The GCDL exam often tests the misconception that cloud adoption is primarily about cost savings or infrastructure migration, when the real transformative capability is automation and AI/ML that fundamentally change business processes and speed.

How to eliminate wrong answers

Option A is wrong because lower storage costs for digitized documents, while beneficial, do not directly reduce processing time from weeks to hours; the bottleneck is manual workflow, not storage cost. Option C is wrong because moving the email system to the cloud has no impact on loan application processing speed; it addresses communication, not core workflow automation. Option D is wrong because migrating from an on-premises Oracle database to Cloud SQL improves database management and scalability but does not automate the manual steps in loan processing; the transformation requires AI/ML and workflow automation, not just a database change.

424
MCQmedium

A team is developing a machine learning model using TensorFlow. They want to train the model on a large dataset stored in Cloud Storage, using GPUs, and then deploy the trained model for online predictions with autoscaling. Which GCP service should they use for the entire workflow?

A.Vertex AI
B.AI Platform (legacy)
C.Cloud Functions
D.Compute Engine with pre-installed ML frameworks
AnswerA

Vertex AI provides a unified managed platform that covers the full ML lifecycle: training with custom or pre-built containers on GPU/TPU, versioning in a Model Registry, and deployment to prediction endpoints with built-in autoscaling based on traffic. It eliminates the need to manually configure infrastructure, allows custom model serving for any framework, and offers MLOps capabilities like monitoring and drift detection. This makes it the recommended service for training and serving TensorFlow models.

Why this answer

Vertex AI is Google Cloud's unified machine learning platform that supports the entire ML workflow: data ingestion from Cloud Storage, training with GPUs, model deployment for online predictions, and autoscaling. It integrates with TensorFlow and provides managed services for each stage. AI Platform (legacy) is deprecated and lacks the full unified experience.

Exam trap

The trap is confusing Vertex AI with legacy AI Platform or assuming that raw Compute Engine is sufficient; the exam tests that Vertex AI is the current, unified service for end-to-end ML workflows.

How to eliminate wrong answers

Option B is wrong because AI Platform (legacy) is an older, deprecated service that does not offer the same integrated workflow or autoscaling capabilities as Vertex AI. Option C is wrong because Cloud Functions is a serverless compute service for event-driven functions, not for training or deploying ML models. Option D is wrong because Compute Engine with pre-installed ML frameworks requires manual setup, scaling, and management, and does not provide a unified workflow for training and deployment.

425
MCQmedium

When choosing a Google Cloud region for a new application, which factors should primarily drive the decision?

A.Always choose `us-central1` because it has the most services and lowest cost.
B.Proximity to users (for low latency), data residency requirements, available services in the region, and pricing.
C.The alphabetical order of region names — 'a' regions are newer and more stable.
D.Google assigns regions automatically based on the user's IP address at account creation.
AnswerB

Region selection is driven by four interdependent factors: network proximity to end users minimizes latency; data residency rules (e.g., GDPR, HIPAA) can mandate storing data in specific geographies; service availability varies because Google rolls out features and specialized hardware (e.g., TPUs, GPUs) gradually across regions; and pricing differences arise from regional energy costs, taxation, and infrastructure overhead. A workload serving Tokyo users should run in an Asian region like asia-northeast1, while an EU workload must consider europe-west regions to satisfy sovereignty requirements.

Why this answer

Selecting a Google Cloud region requires balancing multiple factors: proximity to users minimizes latency for real-time applications; data residency ensures compliance with local regulations (e.g., GDPR); service availability varies by region (e.g., some regions lack GPUs or specific machine types); and pricing differs due to regional operational costs. Google Cloud's global infrastructure is designed to let customers choose regions based on these trade-offs, not on a single criterion.

Exam trap

Google Cloud often tests the misconception that a single 'best' region exists (like us-central1) or that region selection is automated, when in reality it requires a deliberate trade-off analysis of latency, compliance, service availability, and cost.

How to eliminate wrong answers

Option A is wrong because us-central1 does not always have the most services (e.g., some newer services launch first in other regions) and its pricing is not universally the lowest; costs vary by resource type and region. Option C is wrong because alphabetical order has no correlation with region stability or age; Google Cloud regions are named geographically (e.g., us-west1, europe-west4) and stability depends on deployment maturity, not naming. Option D is wrong because Google Cloud does not automatically assign regions based on user IP; region selection is a manual decision made during resource creation, and IP-based assignment would violate customer control over data residency and latency requirements.

426
MCQmedium

A government digital transformation initiative aims to make citizen services available online 24/7. A project manager notes that the technical implementation is proceeding well but citizen adoption remains low. Which dimension of digital transformation has the initiative overlooked?

A.The government should have used a different cloud provider with better uptime guarantees
B.The initiative overlooked user-centered design, accessibility, digital literacy support, and citizen trust-building — critical dimensions of public sector digital transformation beyond technical delivery
C.The government needs to force citizens to use online services by removing all in-person service options
D.The initiative should have started with machine learning features before launching basic services
AnswerB

This captures the overlooked dimensions. Citizens adopt digital services when: the experience is intuitive (user-centered design), accessible to people with disabilities or limited technology experience, accompanied by digital literacy support, and trusted to handle personal data securely. Technical availability is necessary but not sufficient.

Why this answer

Digital transformation in the public sector requires more than just technical deployment; it demands user-centered design, accessibility compliance (e.g., WCAG 2.1), digital literacy programs, and trust-building mechanisms. Without these, even a fully functional cloud-based platform will fail to achieve adoption, as citizens may lack the skills, confidence, or ability to use the service.

Exam trap

Google Cloud often tests the misconception that digital transformation is purely a technology project, leading candidates to focus on cloud providers or advanced AI features instead of the human-centered dimensions like accessibility and trust-building.

How to eliminate wrong answers

Option A is wrong because the issue is not about cloud provider uptime; the technical implementation is already proceeding well, and low adoption stems from human and process factors, not infrastructure reliability. Option C is wrong because forcing citizens to use online services by removing in-person options violates principles of inclusive service delivery and could disenfranchise vulnerable populations, leading to legal and ethical failures. Option D is wrong because starting with machine learning features before basic services would increase complexity and further alienate users who are not yet comfortable with fundamental online interactions.

427
MCQmedium

A company hosts a web application that receives requests from users globally. To handle failures, they run three identical copies of their application behind a load balancer. When one copy fails, the load balancer automatically stops sending traffic to it. What load balancing feature enables this?

A.Round-robin distribution — traffic cycles evenly across all instances regardless of health.
B.Health checks — the load balancer probes backend instances and removes unhealthy ones from the serving pool.
C.SSL termination — decrypting HTTPS traffic before forwarding to backends.
D.Session affinity — routing the same user to the same backend instance.
AnswerB

Health checks are the load balancer's active fault-detection mechanism: it sends periodic probe requests (e.g., HTTP GET, TCP connect, or gRPC health check) to each backend on a configured interval and at a specific path or port. If a backend fails to respond within a timeout or returns an unexpected status code, the load balancer marks it as unhealthy and stops routing new connections to it until it passes the health check again. This continuous probing enables automatic removal of failed instances and ensures traffic is only distributed to backends that can actually serve requests, which is exactly why it is the correct answer.

Why this answer

Health checks are the load balancing feature that proactively monitors the status of backend instances by sending periodic probes (e.g., HTTP GET requests, TCP SYN packets) to a configured endpoint. If a health check fails (e.g., non-2xx response, timeout, or connection refused), the load balancer automatically marks that instance as unhealthy and stops routing new traffic to it, ensuring high availability and fault tolerance.

Exam trap

The GCDL exam often tests the distinction between traffic distribution algorithms (like round-robin) and health monitoring features, leading candidates to mistakenly believe that round-robin inherently handles failures because it 'spreads traffic evenly,' when in fact it has no awareness of instance health.

How to eliminate wrong answers

Option A is wrong because round-robin distribution is a traffic routing algorithm that cycles requests evenly across all instances regardless of their health; it does not detect or react to failures, so it cannot automatically stop sending traffic to a failed copy. Option C is wrong because SSL termination is a feature that offloads the decryption of HTTPS traffic from backend instances to the load balancer; it has no role in monitoring instance health or removing failed instances from the serving pool. Option D is wrong because session affinity (sticky sessions) ensures that requests from the same user are directed to the same backend instance based on a cookie or source IP; it does not provide any failure detection or automatic removal of unhealthy instances.

428
MCQhard

A company uses two different public cloud providers (AWS for their North American operations and Google Cloud for their European operations) to meet data residency requirements and avoid vendor lock-in. Which deployment model does this represent?

A.Hybrid cloud
B.Multi-cloud
C.Multi-region
D.Distributed cloud
AnswerB

Multi-cloud is the deliberate use of two or more distinct public cloud providers—like AWS and Google Cloud—typically to avoid vendor lock-in, achieve geographic coverage, or take advantage of each provider's unique services. Using AWS for North America and Google Cloud for Europe is a classic multi-cloud strategy, as each provider serves a different region while remaining under a unified architectural governance. This approach also provides resilience by ensuring that a failure in one provider does not affect workloads running on the other.

Why this answer

This scenario describes using two distinct public cloud providers (AWS and Google Cloud) to meet data residency and avoid vendor lock-in, which is the definition of a multi-cloud deployment model. Multi-cloud involves using multiple public cloud services from different vendors, as opposed to combining public and private infrastructure (hybrid cloud) or simply deploying across multiple regions within a single provider.

Exam trap

The GCDL exam often tests the distinction between multi-cloud and hybrid cloud, where candidates mistakenly choose hybrid cloud because they confuse 'multiple clouds' with 'mixed public and private infrastructure.'

How to eliminate wrong answers

Option A is wrong because hybrid cloud refers to a mix of private (on-premises) and public cloud infrastructure, not multiple public cloud providers. Option C is wrong because multi-region means deploying resources across multiple geographic regions within a single cloud provider, not across different providers. Option D is wrong because distributed cloud involves a single public cloud provider extending its services to different physical locations (e.g., edge or on-premises), not using multiple independent cloud vendors.

429
MCQeasy

A small business wants to host a static website with minimal management overhead. They want high availability and low cost. Which Google Cloud service should they use?

A.Compute Engine with Apache
B.Cloud Storage with a load balancer
C.App Engine Standard Environment
D.Cloud Run
AnswerB

Cloud Storage with a load balancer is the correct answer because Cloud Storage natively serves static website content (HTML, CSS, JavaScript, images) directly from a bucket without any compute resources, and the load balancer provides a stable HTTPS endpoint, SSL certificate, and global anycast IP. The combination gives you automatic high availability and can be fronted by Cloud CDN for low latency at edge locations, all with zero servers to patch or manage. This is the archetypal 'minimal effort' GCP architecture for a static site: you only upload files and configure the load balancer once.

Why this answer

Cloud Storage with a load balancer is ideal for hosting a static website because it serves content directly from object storage, requires no server management, and the load balancer provides high availability by distributing traffic across multiple regions. This combination offers low cost (pay only for storage and egress) and minimal operational overhead, as there are no virtual machines or application runtimes to maintain.

Exam trap

Google Cloud often tests the misconception that any serverless or managed compute service (like App Engine or Cloud Run) is the best choice for static content, when in fact object storage with a load balancer is simpler and cheaper for purely static assets.

How to eliminate wrong answers

Option A is wrong because Compute Engine with Apache requires managing virtual machines, patching the OS, and configuring Apache, which contradicts the requirement for minimal management overhead and is not cost-effective for a static site. Option C is wrong because App Engine Standard Environment is designed for dynamic web applications with a runtime (e.g., Python, Java) and incurs costs for idle instances, making it overkill and more expensive for a static website. Option D is wrong because Cloud Run is a serverless container platform intended for request-driven applications, not static content; it adds unnecessary complexity and cost compared to directly serving files from Cloud Storage.

430
Multi-Selecthard

A company uses Cloud Monitoring to collect metrics from their applications running on Google Kubernetes Engine (GKE). They want to create custom dashboards and set up alerting policies. Which THREE capabilities are available in Cloud Monitoring? (Choose THREE.)

Select 3 answers
A.Query logs using Logging Query Language
B.Automatically remediate incidents with Cloud Functions
C.Define custom metrics via the Monitoring API
D.Set up alerting policies based on metric thresholds
E.Create uptime checks for external URLs
AnswersC, D, E

The Monitoring API exposes a `timeSeries.create` method that enables you to write custom metrics, such as application-specific counters, gauges, or histograms, into Cloud Monitoring. These custom metrics then appear in dashboards and can be referenced in alerting policies alongside system metrics. This is a core extension point for monitoring anything not automatically collected by Google Cloud's built-in integrations.

Why this answer

The Cloud Monitoring API allows you to define and write custom metrics, which can then be used in dashboards and alerting policies. This is essential for capturing application-specific data that is not automatically collected by the default GKE integration, such as business KPIs or custom performance counters.

Exam trap

The trap here is that candidates confuse Cloud Monitoring with Cloud Logging, mistakenly thinking that log querying (Option A) is a core Monitoring feature, when in fact Monitoring is metric-centric and uses the Metrics Explorer, not the Logs Explorer.

431
MCQhard

A digital media company hosts video content globally. They want to reduce origin server load and deliver content faster to viewers worldwide. Their current architecture routes all viewer requests directly to the origin servers in `us-central1`, causing high latency for viewers in Asia and Europe. Which Google Cloud networking capability addresses this?

A.Deploy identical origin servers in every Google Cloud region globally.
B.Enable Cloud CDN to cache video content at Google's global edge PoPs, serving viewers from the nearest location.
C.Use Cloud VPN to route viewer traffic through a direct tunnel to the origin servers.
D.Increase the origin servers' network bandwidth to handle more simultaneous viewer connections.
AnswerB

Cloud CDN leverages Google's global edge points of presence (PoPs) to cache and serve video content from the location geographically nearest to each viewer. When an Asian viewer requests a video, the request is routed to a nearby edge cache rather than traversing the long-haul network path to us-central1, which dramatically reduces round-trip time and jitter. Additionally, because edge caches absorb the bulk of repeated requests, the origin servers see far fewer direct hits, which reduces origin load and allows the infrastructure to scale cost-effectively for global audiences.

Why this answer

Cloud CDN uses Google's global edge Points of Presence (PoPs) to cache video content closer to viewers, reducing latency and offloading origin servers. When a viewer requests content, Cloud CDN serves it from the nearest edge cache if available, avoiding a direct trip to the origin in us-central1. This directly addresses the high latency for viewers in Asia and Europe without requiring server replication or bandwidth increases.

Exam trap

The GCDL exam often tests the misconception that 'more bandwidth' or 'replicating servers' is the primary solution for global latency, when in fact edge caching (Cloud CDN) is the correct, cost-effective approach for static and dynamic content delivery.

How to eliminate wrong answers

Option A is wrong because deploying identical origin servers in every region is an expensive and operationally complex solution that duplicates infrastructure unnecessarily; Cloud CDN achieves the same latency reduction using caching at edge locations without full server replication. Option C is wrong because Cloud VPN creates an encrypted tunnel for private connectivity between networks but does not cache content or reduce latency for global viewers; it only secures traffic routing, not accelerate delivery. Option D is wrong because increasing origin server bandwidth does not reduce the physical distance between viewers and the server; it only handles more concurrent connections, leaving high latency for distant viewers unresolved.

432
MCQhard

A financial services firm is migrating a legacy application to Google Cloud. The application requires static IP addresses that must not change during migration. The network team wants to minimize downtime. Which migration strategy should they use?

A.Use a global load balancer and update DNS
B.Lift and shift the application to Compute Engine with new IP addresses
C.Use Cloud Interconnect to extend the on-premises network
D.Set up a Cloud VPN tunnel and migrate using the same IP addresses
AnswerD

A Cloud VPN tunnel can extend the on-premises network into Google Cloud by establishing a secure, encrypted connection over the internet. With the tunnel in place, you can advertise the original IP ranges through BGP and create matching VPC subnets, allowing the migrated VM to keep the exact same IP address. During migration, traffic to that IP is routed through the tunnel until cutover, after which you can remove on-prem resources, ensuring a seamless transition without changing any dependent IP references.

Why this answer

A Cloud VPN tunnel can extend the on-premises network into Google Cloud, allowing the legacy application to retain its existing static IP addresses during migration. By establishing a VPN tunnel, traffic can be routed seamlessly between environments, minimizing downtime as the application is migrated without requiring IP address changes.

Exam trap

The trap here is that candidates often confuse connectivity solutions (like Cloud Interconnect or VPN) with IP address preservation, mistakenly thinking that a dedicated connection alone solves the static IP requirement, when in fact the VPN's ability to extend the same subnet is the key enabler.

How to eliminate wrong answers

Option A is wrong because a global load balancer distributes traffic across regions but does not preserve static IP addresses for the application itself; it introduces a new frontend IP and requires DNS changes, which can cause downtime. Option B is wrong because lifting and shifting to Compute Engine with new IP addresses would break the application's dependency on static IPs, requiring reconfiguration and likely causing extended downtime. Option C is wrong because Cloud Interconnect provides a dedicated connection for bandwidth and reliability but does not inherently allow the application to keep its existing IP addresses; it is a connectivity solution, not an IP migration strategy.

433
MCQmedium

A security team is conducting a threat model for their Google Cloud environment. They identify 'insider threat' — a malicious authorized employee who intentionally exfiltrates or destroys data — as a key risk. Which combination of Google Cloud controls most effectively mitigates this risk?

A.Strong external firewall rules, because insider threats come from internal network actors who must be blocked at the perimeter
B.Least privilege IAM (limiting access to only necessary resources), comprehensive audit logging (detecting anomalous access), VPC Service Controls (preventing data exfiltration to external projects), and separation of duties for critical actions
C.Encrypting all data at rest with CMEK, since encryption prevents authorized users from reading data
D.Requiring all employees to pass annual security training to prevent insider threats
AnswerB

This layered approach addresses insider threat from multiple angles: least privilege limits what a malicious insider can access; audit logs detect anomalous behavior (bulk data access, unusual hours); VPC Service Controls prevent copying data to personal or competitor GCP projects; separation of duties requires collusion for the most dangerous actions.

Why this answer

It combines least privilege IAM to limit the blast radius, comprehensive audit logging (e.g., Cloud Audit Logs) to detect anomalous access patterns, VPC Service Controls to prevent data exfiltration via VPC perimeter enforcement, and separation of duties (e.g., using Cloud IAM Conditions) to ensure no single insider can perform critical actions alone. This layered defense addresses both prevention and detection of malicious insider activity.

Exam trap

The trap here is that candidates often assume encryption (CMEK) or training alone can stop insider threats, but they fail to realize that an authorized insider can still read or exfiltrate data unless data exfiltration controls (like VPC Service Controls) and audit logging are in place.

How to eliminate wrong answers

Option A is wrong because strong external firewall rules block external traffic but do not mitigate an insider threat, which originates from within the network and already has valid credentials. Option C is wrong because encrypting data at rest with CMEK does not prevent an authorized user with decryption keys from reading or exfiltrating data; encryption protects against unauthorized access, not insider misuse. Option D is wrong because annual security training is a preventative awareness measure but does not provide technical controls to stop or detect a determined malicious insider who already has access.

434
MCQmedium

An organization wants to protect its web application from DDoS attacks and SQL injection. Which Google Cloud service should they deploy?

A.Cloud DNS
B.Security Command Center
C.Cloud IDS
D.Cloud Armor
AnswerD

Cloud Armor is a security policy service for HTTP(S) load balancers that combines both network-layer DDoS protection and a Web Application Firewall (WAF) with Layer 3-7 filtering. It enforces allow/deny rules at the edge, including preconfigured rules for OWASP Top 10 risks such as SQL injection and cross-site scripting, and scales to absorb volumetric DDoS attacks. Because it operates inline on incoming traffic, it actively blocks malicious requests before they ever reach the application, making it the correct choice for protecting a web application from DDoS and SQL injection.

Why this answer

Cloud Armor provides DDoS protection and Web Application Firewall (WAF) capabilities to block attacks like SQL injection. Cloud IDS is for intrusion detection. Security Command Center is for vulnerability management.

Cloud DNS is for DNS resolution.

435
MCQeasy

What compliance certification verifies that an organization's Information Security Management System (ISMS) meets internationally recognized standards for managing information security risks?

A.SOC 2 Type II
B.ISO/IEC 27001
C.PCI DSS
D.FedRAMP
AnswerB

ISO/IEC 27001 is an internationally recognized standard that specifies the requirements for establishing, implementing, maintaining, and continually improving an information security management system (ISMS). It is a voluntary certification standard, and Google Cloud has achieved ISO/IEC 27001 certification through independent third-party audits, demonstrating alignment with global best practices for information security management. This certification provides customers with assurance that Google Cloud has systematic, risk-based processes for managing sensitive information, making it the correct answer.

Why this answer

ISO/IEC 27001 is the international standard that specifies the requirements for establishing, implementing, maintaining, and continually improving an Information Security Management System (ISMS). It provides a systematic approach to managing sensitive company information, ensuring it remains secure through risk management processes. This certification is recognized globally and is the primary standard for ISMS compliance.

Exam trap

The trap here is that candidates often confuse SOC 2 Type II (which focuses on service organization controls) with ISO/IEC 27001 (which is the specific international standard for an ISMS), leading them to select SOC 2 Type II when the question explicitly asks for an ISMS certification.

How to eliminate wrong answers

Option A is wrong because SOC 2 Type II is an auditing procedure that evaluates a service organization's controls related to security, availability, processing integrity, confidentiality, and privacy based on the AICPA Trust Services Criteria, not an ISMS standard. Option C is wrong because PCI DSS is a security standard specifically for organizations that handle branded credit cards, focusing on cardholder data protection, not a general ISMS framework. Option D is wrong because FedRAMP is a U.S. government program that provides a standardized approach to security assessment, authorization, and continuous monitoring for cloud products and services, not an international ISMS certification.

436
MCQeasy

A company wants to enforce that all Cloud Storage buckets in a project have uniform bucket-level access enabled. Which Google Cloud tool should they use?

A.Use Cloud Audit Logs to monitor and alert on non-compliant buckets.
B.Define an Organization Policy with a constraint on uniform bucket-level access.
C.Set an IAM policy to deny access to buckets without uniform access.
D.Use Cloud Key Management Service to rotate keys.
AnswerB

An Organization Policy with the constraint *storage.uniformBucketLevelAccess* enforces that every bucket in a given hierarchy must have uniform bucket-level access enabled. When set to True, this list constraint blocks any request to create or update a bucket that does not have uniform access, effectively acting as an immutable compliance guardrail. This is the correct because it is a prerequisite enforcement method provided by Google Cloud's organization policy service.

Why this answer

Organization Policies in Google Cloud allow administrators to enforce constraints across the entire resource hierarchy. The constraint `constraints/storage.uniformBucketLevelAccess` can be applied at the project, folder, or organization level to require uniform bucket-level access on all Cloud Storage buckets, preventing any bucket from being created or updated without it.

Exam trap

Google Cloud often tests the distinction between monitoring/logging tools (like Audit Logs) and enforcement tools (like Organization Policies), leading candidates to choose a reactive solution instead of a proactive, policy-based one.

How to eliminate wrong answers

Option A is wrong because Cloud Audit Logs only provide logging and monitoring for historical or real-time events; they cannot enforce or prevent non-compliant configurations, only alert after the fact. Option C is wrong because IAM policies grant or deny access to resources based on identities and roles, but they cannot enforce a configuration setting like uniform bucket-level access on a bucket itself. Option D is wrong because Cloud Key Management Service (KMS) manages encryption keys, not access control policies for bucket-level access settings.

437
MCQeasy

A company wants to migrate its on-premises MySQL database to Google Cloud with minimal changes to the application. Which managed database service should they use?

A.Cloud Spanner
B.Cloud Bigtable
C.Cloud SQL for MySQL
D.Firestore
AnswerC

Cloud SQL for MySQL is Google Cloud's fully managed service that is natively compatible with the MySQL engine, including support for MySQL 5.7, 8.0, and 8.4. It provides the same SQL syntax, stored procedures, triggers, and InnoDB storage engine, so existing applications can connect with standard MySQL drivers and require minimal to no code changes. With automated backups, high availability, read replicas, and straightforward import tools, it is the correct and lowest-risk choice for migrating an on-premises MySQL database.

Why this answer

Cloud SQL provides managed MySQL, PostgreSQL, and SQL Server databases. It is compatible with existing MySQL applications. Cloud Spanner is globally distributed but not drop-in MySQL.

Firestore and Bigtable are NoSQL.

438
MCQhard

A company wants to migrate its on-premises Oracle database to Google Cloud. They need PostgreSQL compatibility with high performance for transaction processing and built-in support for AI-driven optimisations. Which database service should they choose?

A.Cloud Spanner
B.Bigtable
C.Cloud SQL for PostgreSQL
D.AlloyDB
AnswerD

AlloyDB is a fully managed, PostgreSQL-compatible database service engineered for high performance and scalability, with AI-driven optimizations such as adaptive caching and an integrated columnar engine. It provides up to 4x faster transactional performance and 10x faster analytical queries than standard PostgreSQL, and its compatibility layer supports Oracle-like data types, functions, and SQL syntax, easing migration from Oracle. These capabilities directly align with the company's requirement for a PostgreSQL-compatible database with intelligent performance enhancements.

Why this answer

AlloyDB is Google Cloud's fully managed PostgreSQL-compatible database designed for high-performance transactional workloads, with built-in AI-driven optimizations such as adaptive indexing and automatic query tuning. It offers superior performance compared to standard Cloud SQL for PostgreSQL while maintaining full PostgreSQL compatibility.

Exam trap

The trap is confusing Cloud SQL for PostgreSQL with AlloyDB — both are PostgreSQL-compatible, but only AlloyDB provides the high-performance, AI-driven optimizations described in the question.

How to eliminate wrong answers

Option A is wrong because Cloud Spanner is a globally distributed, horizontally scalable relational database but is not PostgreSQL-compatible in the traditional sense and is not optimized for AI-driven optimizations. Option B is wrong because Bigtable is a NoSQL wide-column store, not a relational PostgreSQL-compatible database. Option C is wrong because Cloud SQL for PostgreSQL is PostgreSQL-compatible but does not provide the high-performance, AI-driven optimizations that AlloyDB offers for demanding transactional workloads.

439
Multi-Selecthard

A company runs a batch processing job every hour using Cloud Dataflow. They notice increasing costs and want to optimize. Which three actions would reduce cost? (Choose exactly 3.)

Select 3 answers
A.Use preemptible VMs for worker nodes
B.Shut down the Dataflow job between runs
C.Switch from batch to streaming mode
D.Set autoscaling to a lower maximum number of workers
E.Use flexible resource scheduling (batch mode)
AnswersA, D, E

Preemptible VMs (now called spot VMs in Google Cloud) provide a significantly discounted price (typically 60-80% off on-demand pricing) for worker nodes in a Dataflow batch pipeline. Because Dataflow is designed to handle worker loss through checkpointing and automatic restart of tasks, batch jobs are generally resilient to the occasional termination that preemptible VMs may undergo. This directly reduces the compute cost of the worker pool without changing the pipeline logic or delivery time, making it the most straightforward way to cut costs for a recurring batch job.

Why this answer

Option A is correct because Dataflow batch jobs can run on preemptible VMs (now called Spot VMs), which cost substantially less than standard Compute Engine instances; Dataflow handles preemption by rescheduling work, making this a standard cost optimization for batch pipelines. Option D is correct because capping the autoscaling maximum number of workers prevents the job from over-provisioning workers during demand spikes, directly limiting the compute cost incurred per run. Option E is correct because Flexible Resource Scheduling (FlexRS) in batch mode uses a mix of preemptible and standard VMs with a delayed, lower-cost scheduling window, reducing the effective cost of batch processing.

Option B is not appropriate because a batch Dataflow job terminates on its own when the pipeline finishes, so there is no running job to shut down between hourly runs. Option C is incorrect because switching from batch to streaming keeps workers running continuously and typically increases cost, the opposite of the goal.

Exam trap

GCDL often tests the misconception that streaming mode is cheaper or that stopping jobs saves money — candidates pick 'switch to streaming' or 'shut down between runs' without realizing streaming is more expensive and batch jobs already terminate.

440
Matchingmedium

Match each Google Cloud storage class to its use case.

Drag a concept onto its matching description — or click a concept then click the description.

Concepts
Matches

Frequently accessed data, low latency

Data accessed less than once a month

Data accessed less than once a quarter

Data accessed less than once a year

Automatic placement of objects into appropriate classes

Why these pairings

The correct matches are: Standard for frequently accessed data, Nearline for data accessed less than once a month, Coldline for data accessed less than once a quarter, and Archive for data accessed less than once a year. Common confusions involve misremembering the access frequency thresholds.

441
MCQhard

A security team needs to monitor and analyze logs from multiple GCP projects to detect threats across the organization. They require a SIEM solution that can ingest logs from on-premises and other clouds. Which service should they use?

A.Security Command Center
B.Cloud IDS
C.Chronicle
D.Cloud Logging
AnswerC

Chronicle is a cloud-native SIEM from Google Cloud designed to ingest, normalize, and analyze security logs from multiple cloud providers, on-premises systems, and SaaS applications. Its architecture leverages BigQuery for scalable log storage and provides advanced threat detection, retroactive analysis, and custom detection rules with low-latency searching. As a Google Cloud product purpose-built for security log analytics, it directly addresses the need to monitor and analyze logs across a multi-cloud environment.

Why this answer

Chronicle is a Google Cloud SIEM that ingests logs from various sources, including on-premises and other clouds, and provides threat detection. Security Command Center is for vulnerability scanning, not SIEM.

442
MCQeasy

A company wants to monitor the CPU and memory utilisation of their Compute Engine instances and set up alerts when utilisation exceeds 80%. Which Google Cloud service should they use?

A.Cloud Monitoring
B.Cloud Logging
C.Cloud Error Reporting
D.Cloud Trace
AnswerA

Cloud Monitoring is the correct service because it is purpose-built to ingest and store numeric time-series metrics, such as CPU utilization and memory usage, from resources like Compute Engine and Kubernetes. It provides alerting policies that trigger notifications when these metrics cross user-defined thresholds, and offers dashboards and querying via MQL or PromQL. This is the standard GCP tool for real-time infrastructure observability, not just logging or error tracking.

Why this answer

Cloud Monitoring collects metrics from Compute Engine instances via the Ops Agent (or legacy monitoring agent) and provides alerting policies based on metric thresholds. CPU utilization and memory utilization are standard metrics (compute.googleapis.com/instance/cpu/utilization and memory/utilization) that can be charted and alerted on when exceeding 80%. This is the native GCP service for metric-based monitoring and alerting.

Exam trap

The trap is confusing monitoring with logging or tracing — candidates may pick Cloud Logging because they think alerts come from logs, but metric-based threshold alerts belong to Cloud Monitoring.

How to eliminate wrong answers

Option B is wrong because Cloud Logging is for ingesting, storing, and analyzing log data, not for metric-based threshold alerting on resource utilization. Option C is wrong because Cloud Error Reporting aggregates and displays application errors from logs, not infrastructure metrics. Option D is wrong because Cloud Trace is a distributed tracing system for latency analysis of requests, not resource utilization monitoring.

443
MCQmedium

A team runs a production Compute Engine instance that has been running for 15 days in a 30-day month. They also have a second instance that runs occasionally for testing. They want to maximize cost savings without committing to a 1-year or 3-year term. Which discount will apply automatically?

A.Committed use discount (CUD)
B.Sustained use discount (SUD)
C.Preemptible VM discount
D.No discount applies automatically.
AnswerB

Sustained use discounts (SUDs) are automatically applied to standard Compute Engine instances based on how long each instance runs within a billing month, with no sign-up, plan, or upfront commitment required. Once an instance exceeds 25% of the month (more than about 7.5 days in a 30-day month), the discount kicks in automatically; a production instance running 15 days clearly crosses that threshold. The discount scales with usage and can reach up to 30% off the on-demand hourly rate for the instance when it runs for the entire month, making SUD the correct fit here.

Why this answer

Sustained use discounts automatically apply to instances that run for more than 25% of a month (i.e., >7.5 days). At 15 days, the instance qualifies for sustained use discounts. Committed use discounts require a pre-purchased commitment.

444
MCQmedium

A company uses Cloud SQL and wants to encrypt data at rest with a key that they manage and rotate themselves. They also want to ensure that the encryption happens automatically before data is written to disk. Which configuration should they choose?

A.Enable default Google-managed encryption
B.Store the encryption key in Secret Manager and configure Cloud SQL to use it
C.Use a Customer-Managed Encryption Key (CMEK) from Cloud KMS
D.Use a Customer-Supplied Encryption Key (CSEK)
AnswerC

CMEK in Cloud KMS lets you create and control a key that Cloud SQL automatically uses for encrypting data, backups, and logs at rest. You control key rotation, enabling/disabling, and access via IAM, and Cloud KMS records key usage for auditing. This gives the required customer control while keeping transparent, automatic encryption.

Why this answer

Customer-Managed Encryption Keys (CMEK) in Cloud KMS let you own, manage, and rotate the key while Cloud SQL automatically uses it to encrypt data at rest before writing to disk. This satisfies both the customer-managed rotation requirement and the automatic pre-write encryption requirement.

Exam trap

GCDL often tests the confusion between CMEK and CSEK — candidates pick CSEK thinking 'customer-supplied' means customer-managed, but CSEK is client-provided per-request and does not provide automatic at-rest encryption managed by the service.

How to eliminate wrong answers

Option A is wrong because Google-managed encryption uses keys owned and rotated by Google, not the customer, so it fails the customer-managed requirement. Option B is wrong because storing a key in Secret Manager does not integrate with Cloud SQL's at-rest encryption — Secret Manager is for application secrets, not for CMEK integration. Option D is wrong because Customer-Supplied Encryption Keys (CSEK) are passed by the client at request time and are not persisted by Google, so they do not provide automatic at-rest encryption managed by the service — CSEK is also not supported for Cloud SQL in the same way as CMEK.

445
MCQeasy

When a company moves from maintaining its own data center to using Google Cloud, which operational responsibility does Google assume that the company previously managed?

A.Writing and maintaining application code
B.Physical hardware maintenance, data center facilities, and network equipment management
C.Defining which users can access the company's applications
D.Backing up the company's application data
AnswerB

Google assumes full ownership of the physical layer in its data centers, including server hardware, rack layout, cooling systems, electrical power, and networking equipment. The company also manages physical access controls, biometric security, and on-site staff. For a customer migrating from on-premises data centers, this burden no longer falls on their facilities and engineering teams, which is a primary cost and complexity reduction of moving to cloud.

Why this answer

When a company migrates from an on-premises data center to Google Cloud, Google assumes responsibility for the physical infrastructure, including hardware maintenance, facility management (power, cooling, security), and network equipment. This is the core of the cloud provider's shared responsibility model, where the provider manages the 'cloud' while the customer manages what is 'in' the cloud. Option B correctly identifies these operational responsibilities that shift to Google.

Exam trap

The GCDL exam often tests the shared responsibility model by making candidates confuse which responsibilities shift to the cloud provider versus those that remain with the customer, especially by implying that the provider handles all security or data management tasks, when in fact the customer retains control over access, data, and application logic.

How to eliminate wrong answers

Option A is wrong because writing and maintaining application code remains the customer's responsibility under the shared responsibility model; Google Cloud provides the platform but does not develop or maintain customer applications. Option C is wrong because defining user access to applications is a customer-managed identity and access management (IAM) task, even though Google Cloud provides IAM tools, the customer must configure and enforce policies. Option D is wrong because while Google Cloud offers backup services (e.g., Cloud Storage, snapshots), the customer is responsible for configuring, scheduling, and verifying backups of their application data; Google does not automatically back up customer data unless explicitly configured by the customer.

446
MCQmedium

A retail company needs to process financial transactions requiring strict ACID compliance, serve global customers with consistent low-latency reads and writes, and scale horizontally without downtime for maintenance. Which Google Cloud database service is uniquely designed to meet all three requirements simultaneously?

A.Cloud SQL with read replicas in multiple regions
B.Cloud Spanner, Google's globally distributed relational database with ACID transactions and horizontal scalability
C.Firestore in Datastore mode with multi-region replication
D.BigQuery with streaming inserts for real-time transaction processing
AnswerB

Cloud Spanner uniquely satisfies all three requirements: full ACID compliance for financial transactions, global distribution with strong consistency for low-latency global reads and writes, and horizontal scaling without downtime. It was purpose-built by Google for exactly this class of globally consistent, highly available transactional workloads.

Why this answer

Cloud Spanner is the only Google Cloud database service that simultaneously provides ACID compliance across global transactions, consistent low-latency reads and writes via TrueTime and synchronous replication, and horizontal scaling without downtime through automatic sharding and resharding. It uniquely combines relational database semantics with global distribution, making it the correct choice for the given requirements.

Exam trap

Google Cloud often tests the misconception that a globally distributed NoSQL database like Firestore can provide ACID compliance, but Firestore only supports single-entity transactions and lacks the strict consistency needed for financial transactions.

How to eliminate wrong answers

Option A is wrong because Cloud SQL with read replicas does not support horizontal scaling for writes (it is a single-writer, multi-reader architecture) and cannot achieve consistent low-latency writes across global regions without downtime for maintenance. Option C is wrong because Firestore in Datastore mode is a NoSQL database that does not support ACID transactions across multiple entities (only single-entity transactions are atomic) and cannot provide strict ACID compliance for financial transactions. Option D is wrong because BigQuery is an analytical data warehouse designed for large-scale queries, not for transactional processing with ACID guarantees, and its streaming inserts are eventually consistent and not suitable for real-time transaction processing.

447
MCQhard

Google Cloud's infrastructure is designed to be highly available across multiple failure domains. What are 'availability zones' in Google Cloud, and how do they differ from 'regions'?

A.Zones are continents; regions are individual countries within a continent.
B.A region is a geographic area containing multiple isolated zones; zones have independent failure domains but low-latency connectivity within the region.
C.Zones and regions are different terms for the same thing — Google uses them interchangeably.
D.A zone is a global resource; a region is a local data center.
AnswerB

A GCP region is a geographic area, typically a city or metropolitan area, that hosts at least three zones, each a physically separate data center with independent power, cooling, and network connectivity. These zones are isolated failure domains: an outage in one zone does not impair the others, which is the foundation for building high availability within a single region. At the same time, zones in a region are interconnected by low-latency links (commonly under 5 ms round trip), enabling synchronous replication and active-active workload designs across zones without cross-country delays. Thus, the combination of isolation and low latency lets you survive zone failures while maintaining fast performance.

Why this answer

In Google Cloud, a region is a specific geographic location composed of multiple zones, each of which is an isolated failure domain with independent power, cooling, and networking. Zones within the same region are connected by low-latency, high-bandwidth links, enabling high availability and fault tolerance for applications. This design ensures that a failure in one zone does not affect resources in another zone within the same region.

Exam trap

The trap here is that candidates often confuse zones with regions, thinking they are synonymous or hierarchical in a simplistic way (e.g., zones as sub-regions), rather than understanding that zones are independent failure domains within a region with low-latency interconnects.

How to eliminate wrong answers

Option A is wrong because zones are not continents; they are discrete data center clusters within a region, and regions are not individual countries but broader geographic areas that may span multiple countries or states. Option C is wrong because zones and regions are distinct concepts in Google Cloud; they are not interchangeable terms, and using them as such would lead to incorrect architectural decisions. Option D is wrong because a zone is not a global resource; it is a local deployment area within a region, and a region is not a single local data center but a collection of zones.

448
MCQeasy

Which Google Cloud service is a managed platform for building, training, and deploying ML models, including support for AutoML and custom models?

A.BigQuery ML
B.AI Platform (legacy)
C.Cloud TPU
D.Vertex AI
AnswerD

Vertex AI is Google Cloud's unified, end-to-end managed ML platform that covers the full model lifecycle, from data preparation and feature engineering to AutoML or custom training, hyperparameter tuning, model validation, deployment, and continuous monitoring. It provides a single API and workflow that integrates with Cloud Storage, BigQuery, and other Google Cloud services, while also offering advanced MLOps components like Vertex AI Pipelines, Model Registry, and Vertex AI Feature Store. As a fully managed service, Vertex AI abstracts infrastructure management, enabling automated autoscaling for prediction endpoints, security policies, and versioning, making it the correct answer to the question.

Why this answer

Vertex AI is a unified ML platform that combines AutoML and custom model training, tuning, and serving.

449
Multi-Selectmedium

A company needs to organize their GCP resources into a hierarchy that reflects their departments: Engineering, Marketing, and Finance. Each department has multiple projects. They also want to apply common policies to all departments except Finance, which has special compliance requirements. Which TWO steps should they take?

Select 2 answers
A.Use labels to differentiate departments instead of folders.
B.Apply organization policies at the organization node and override them for the Finance folder using tags.
C.Apply separate IAM policies for each project within a folder.
D.Create a project for each department and place resources in it.
E.Create a folder for each department.
AnswersB, E

Organization policies set at the organization node are inherited by all descendants, establishing a consistent security and compliance baseline. However, using tags as condition keys in policy constraints allows you to create exceptions, such as permitting a specific resource setting only for the Finance folder. This pattern preserves centralized control while accommodating department-specific needs, aligning with the recommended practice of least privilege and scoped flexibility.

Why this answer

Option E is correct because GCP resource hierarchy uses folders to model organizational structure such as departments, so creating a folder for Engineering, Marketing, and Finance lets each department contain its multiple projects and inherit policies from above. Option B is correct because organization policies are inherited down the hierarchy, so applying common policies at the organization node enforces them across all departments, while tags on the Finance folder allow a conditional override to satisfy its special compliance requirements. Option A is not appropriate because labels are for metadata, billing, and filtering, not for building a policy-inheritance hierarchy.

Option C is wrong because setting separate IAM policies per project does not create the required departmental hierarchy or centralized policy inheritance. Option D is wrong because a single project per department cannot hold multiple projects and does not reflect the folder-based hierarchy needed for policy inheritance.

Exam trap

The trap is thinking labels can replace folders for policy inheritance; labels are for metadata and billing, not for hierarchical policy enforcement.

450
MCQeasy

A retail company experiences sudden traffic spikes during flash sales. Their on-premises infrastructure often runs out of capacity, causing website slowdowns. Which cloud characteristic directly solves this problem?

A.Pay-as-you-go pricing
B.Scalability
C.Global reach
D.Security
AnswerB

Scalability is the property that enables a system to handle growing workload by adding or reallocating resources, and in cloud environments it is implemented through autoscaling groups, load balancers, and elastic capacity pools. When traffic spikes, the scaling policy can launch additional virtual machines or containers within minutes, ensuring that the application continues to respond despite the sudden load. This is precisely what the retail company needs to maintain availability and performance during promotional events or flash sales.

Why this answer

Scalability is the cloud characteristic that allows infrastructure to automatically expand or contract in response to demand, which directly addresses the flash-sale traffic spikes that overwhelm fixed on-premises capacity. In cloud environments, horizontal scaling (adding more instances) or vertical scaling (larger instances) can be triggered by metrics like CPU or request rate, so the website absorbs the surge without slowdowns. Pay-as-you-go pricing is a billing model, not a capacity mechanism, so it does not solve the capacity shortfall itself.

Exam trap

The trap here is confusing pay-as-you-go pricing (a cost model) with scalability (a capacity model); candidates often pick the billing answer because it sounds like the 'cloud advantage' they memorized, but the question asks what solves the capacity problem.

How to eliminate wrong answers

Option A is wrong because pay-as-you-go pricing only changes how you are billed for consumed resources; it does not add capacity when traffic spikes, so the site would still slow down. Option C is wrong because global reach refers to deploying resources in multiple geographic regions to reduce latency for distributed users, not to dynamically handling a sudden capacity surge. Option D is wrong because security (IAM, encryption, network controls) protects data and access but has no role in provisioning additional compute or bandwidth during a traffic spike.

Page 5

Page 6 of 12

Page 7