Examkin

All practice questions

1598 questions

All types1,598Multiple choice815Multiple select537Numeric8Matching64Ordering174
Question 221Question →

A smart building automation company processes real-time telemetry from thousands of commercial IoT gateways. The application runs in Google Cloud with primary compute and data resources deployed in region `us-central1`, and a Warm Standby disaster recovery (DR) setup configured in region `us-east4`. The company specifies a Recovery Time Objective (RTO) of 15 minutes and a Recovery Point Objective (RPO) of 1 minute. During an unannounced disaster recovery failover drill of `us-central1`, the operational team promoted the secondary database replica, but the DR secondary Managed Instance Group (MIG) in `us-east4` failed to scale out to accommodate production traffic due to compute capacity errors. Which architectural modification should the Cloud Architect implement to ensure seamless disaster recovery execution?

Show answer & explanation

Answer: Proactively request and maintain sufficient Compute Engine regional resource quotas in `us-east4` to match primary production peak requirements.

Answer

Proactively request and maintain sufficient Compute Engine regional resource quotas in `us-east4` to match primary production peak requirements.
When executing a disaster recovery failover to a secondary region, compute resource quotas must be requested and maintained in advance. If quotas are not pre-allocated, Compute Engine Managed Instance Groups will fail to scale up, preventing the application from reaching operational capacity within the target Recovery Time Objective (RTO).

Step-by-Step Solution

1
Analyze the failure mode during the regional failover drill.
Identified that compute instance provisioning failed because the target region lacked adequate vCPU or instance quotas for the scaled workload.
Managed Instance Groups cannot scale beyond the quota limits assigned to the target region.
2
Evaluate disaster recovery capacity planning principles.
Determined that regional quotas must be requested and approved in advance of any operational incident.
Quota increases require evaluation and cannot be guaranteed instantly during an emergency.
3
Select the correct mitigation strategy to meet the 15-minute RTO requirement.
Maintaining pre-approved quotas in the secondary DR region ensures immediate scale-out capability during failover.
Pre-allocated quotas eliminate provisioning blocks and fulfill strict RTO constraints.

Key Concept

Disaster Recovery Capacity Planning and Regional Quotas
Question 222Question →

A global financial intelligence firm is deploying a hybrid cloud architecture to connect its primary data center to Google Cloud. The workload requires a dedicated network link providing 15 Gbps of sustained throughput with predictable low latency to a central network hub. Additionally, multiple project teams require secure, isolated network environments that can directly access on-premises databases without traversing public internet endpoints. Which TWO architectural decisions should the lead cloud architect implement to meet these requirements? (Select TWO)

Select all that apply

Show answer & explanation

Answer: Provision Dedicated Interconnect connections with Cloud Router dynamic BGP routing to satisfy the throughput and low-latency requirements.; Configure a Shared VPC architecture where isolated teams deploy workloads into service projects attached to the host VPC containing the hybrid interconnect.

Answer

The architect must provision Dedicated Interconnect with Cloud Router BGP routing to support 15 Gbps throughput and configure a Shared VPC network with service projects to grant isolated team workloads direct access to on-premises resources.
The combination of Dedicated Interconnect with Cloud Router dynamic BGP routing and a Shared VPC topology correctly satisfies all requirements. Dedicated Interconnect handles high-throughput (15 Gbps) requirements with low latency over private circuits. Shared VPC provides centralized control over hybrid connectivity while enabling isolated service projects to securely access on-premises databases directly through the host network without relying on transitive VPC peering.

Step-by-Step Solution

1
Evaluate hybrid connectivity bandwidth and performance requirements.
Identified that 15 Gbps sustained throughput exceeds Cloud VPN maximum single-tunnel capacities (3 Gbps) and requires Dedicated Interconnect (10 Gbps or 100 Gbps circuits).
Dedicated Interconnect delivers physical private circuits suitable for high-bandwidth enterprise traffic requiring SLA-backed performance.
2
Evaluate cross-project network isolation and hybrid access constraints.
Determined that Shared VPC allows service project workloads to directly use host VPC subnet routes leading to on-premises networks.
VPC Network Peering does not support transitive routing from on-premises to peered spokes, whereas Shared VPC natively exposes host connectivity to service projects.

Key Concept

Designing High-Throughput Hybrid Connectivity and Shared VPC Architectures
Question 223Question →

A fintech enterprise is modernizing a legacy online payment portal to eliminate architectural technical debt during its migration from an on-premises data center to Google Cloud. The application architecture consists of lightweight, stateless HTTP API microservices and a standard single-region 3 TB relational database. The migration strategy mandates minimizing operational management overhead for compute while maintaining backward compatibility for database schema updates to support continuous rollout without downtime.

Which TWO architectural actions should the team take to achieve these requirements? (Select TWO)

Select all that apply

Show answer & explanation

Answer: Deploy the stateless HTTP API microservices to Cloud Run.; Migrate the database to Cloud SQL using Database Migration Service (DMS) and implement an expand-and-contract schema pattern during migration.

Answer

The team should deploy the stateless HTTP microservices to Cloud Run and migrate the relational database to Cloud SQL using Database Migration Service (DMS) while leveraging an expand-and-contract schema pattern.
Deploying stateless microservices to Cloud Run minimizes operational management by using a fully managed serverless execution model. Migrating the relational database to Cloud SQL using Database Migration Service provides a streamlined migration path, while using an expand-and-contract schema pattern ensures zero-downtime schema migrations that maintain backward compatibility across old and new application versions.

Step-by-Step Solution

1
Evaluate compute modernization choices against technical debt and operational overhead constraints.
Cloud Run provides a fully managed serverless compute environment for containerized stateless HTTP microservices without requiring node provisioning or cluster control plane management.
Choosing GKE for simple stateless services creates unnecessary operational complexity and maintenance burden.
2
Select database destination and continuous replication migration tooling.
Cloud SQL provides managed relational database capabilities suitable for single-region workload sizes, managed seamlessly via Database Migration Service (DMS).
Cloud Spanner is designed for globally distributed scale and adds unnecessary cost and refactoring overhead for standard single-region relational databases.
3
Define database schema migration strategy to ensure zero-downtime rollout.
The expand-and-contract pattern allows new and old application versions to read/write concurrently during deployment by making non-breaking incremental schema changes.
Executing breaking database schema changes during live migration disrupts legacy application instances.

Key Concept

Managing Technology Debt and Legacy Cloud Migrations
Question 224Question →

A pharmaceutical distribution company manages its Google Cloud infrastructure using Terraform within a centralized Cloud Build CI/CD pipeline. During an emergency operational incident, an engineer manually updated the machine type and auto-scaling limits of an unmanaged instance group directly in the Google Cloud Console. The infrastructure team now needs to resolve this configuration drift and bring the state back into alignment with IaC governance standards without tearing down existing production workloads. Which approach should the cloud architect recommend?

Show answer & explanation

Answer: Update the local Terraform HCL codebase to reflect the upgraded instance group specifications, execute a terraform plan to confirm zero resource destruction, and commit the changes to trigger the pipeline.

Answer

Update the local Terraform HCL codebase to reflect the upgraded instance group specifications, execute a terraform plan to confirm zero resource destruction, and commit the changes to trigger the pipeline.
The correct approach is to update the Terraform HCL code to match the manual configuration changes, perform a plan check to ensure no destructive updates are scheduled, and commit the code to version control. This restores parity between code, state, and live GCP resources while maintaining IaC governance.

Step-by-Step Solution

1
Identify the drift source
Recognize that manual modifications performed in the Cloud Console caused the live environment to deviate from the stored state file.
Infrastructure as Code requires declarative code to represent the true desired state of managed cloud resources.
2
Update Terraform code definition
Modify the instance group resource block parameters in HCL to match the updated live resource attributes.
Bringing code in line with manual emergency changes captures valid operational adjustments in version control.
3
Validate state execution plan
Run terraform plan to verify that Terraform detects no unexpected changes or resource replacements.
Prevents accidental downtime or recreation of production resources before committing to main branches.

Key Concept

Remediating Infrastructure Configuration Drift in Terraform
Estimated Time:1m 30s
Question 225Question →

An online multiplayer gaming platform is architecting its global matchmaking and player profile backend on Google Cloud. The architecture must achieve high availability across regional outages with a Recovery Point Objective (RPO) of 0 and a Recovery Time Objective (RTO) of less than 1 minute for transactional player state. Additionally, the compute tier handles persistent WebSocket connections and I/O-intensive queue processing, which does not correlate directly with CPU usage. Which TWO architectural decisions should the Cloud Architect implement to satisfy these technical requirements? (Select 2 answers)

Select all that apply

Show answer & explanation

Answer: Deploy a multi-region Cloud Spanner instance to provide globally distributed strongly consistent transactions with zero RPO across regional failures.; Configure Kubernetes Horizontal Pod Autoscaler (HPA) using custom metrics for active WebSocket connections and queue length rather than default CPU utilization.

Answer

The architect should deploy a multi-region Cloud Spanner instance for zero-RPO transactional data and configure Kubernetes Horizontal Pod Autoscaler using custom metrics for active connections and queue length.
Deploying a multi-region Cloud Spanner instance guarantees synchronous multi-region replication with zero data loss (RPO = 0) and automatic sub-minute failover. Additionally, configuring HPA with custom application metrics (WebSocket connections and queue depth) ensures the compute tier scales appropriately for I/O-bound workloads that do not correlate with CPU consumption.

Step-by-Step Solution

1
Evaluate storage tier against RPO=0 and sub-minute RTO requirements during regional failures.
Multi-region Cloud Spanner provides synchronous replication across regions, meeting RPO=0 and sub-minute failover, whereas asynchronous Cloud SQL replicas risk data loss and slower failover.
Cloud Spanner uses Paxos consensus across multi-region zones to guarantee strong consistency with minimal recovery objectives.
2
Evaluate autoscaling criteria for persistent connection and queue-bound microservices.
Scaling based on custom metrics (active WebSocket connections and queue depth) accurately matches workload demand, whereas CPU metrics lag or misjudge I/O bottlenecks.
Connection-heavy workloads consume memory and network sockets without spiking CPU, making CPU-based autoscaling ineffective.

Key Concept

Designing multi-region database architectures for zero RPO/RTO and autoscaling I/O-bound compute workloads using custom metrics.
Question 226Question →

An online learning platform automates its Google Cloud infrastructure deployments using Terraform executed within a continuous integration and continuous delivery (CI/CD) pipeline. The Lead Cloud Architect must establish a secure governance model for deployment pipeline authentication, permission scoping, and Terraform state management following Google Cloud best practices. Which design strategy should the architect implement?

Show answer & explanation

Answer: Authenticate the CI/CD pipeline using Workload Identity Federation, grant minimal predefined IAM roles to the deployment service account, and store state files in a Cloud Storage bucket with object versioning and state locking enabled.

Answer

Authenticate the CI/CD pipeline using Workload Identity Federation, grant minimal predefined IAM roles to the deployment service account, and store state files in a Cloud Storage bucket with object versioning and state locking enabled.
Google Cloud best practices for Infrastructure as Code (IaC) pipelines require keyless authentication via Workload Identity Federation, strict scoping of deployment service accounts using fine-grained predefined IAM roles, and centralized remote state management in Cloud Storage with state locking and versioning enabled.

Step-by-Step Solution

1
Configure pipeline authentication without service account keys
Workload Identity Federation allows external CI/CD workloads to exchange short-lived tokens for GCP IAM credentials securely.
Eliminating exportable JSON keys prevents credential exfiltration and management toil.
2
Enforce the principle of least privilege for deployment permissions
Assign specific, predefined IAM roles required only for the target resources managed by the pipeline.
Avoids over-privileged primitive roles like Owner or Editor that create broad security risks.
3
Configure a centralized, secure remote state backend
Store Terraform state in a Cloud Storage bucket configured with state locking and object versioning.
Prevents concurrent execution state corruption and enables recovery from bad state updates.

Key Concept

IaC Security Governance and Remote State Management
Question 227Question →

A digital media processing company is migrating its video catalog workflow to Google Cloud. The architecture includes batch-oriented, stateless video rendering tasks that run intermittently when new media arrives, as well as a 600 TB600\text{ TB} historical video archive that is retained for compliance and accessed less than once a year. The business objective is to minimize ongoing infrastructure maintenance overhead and optimize total operational expenditure. Which TWO architectural decisions should the cloud architect recommend? (Select TWO)

Select all that apply

Show answer & explanation

Answer: Execute the batch video rendering tasks using Cloud Run Jobs to eliminate baseline compute costs and automatically scale to zero when idle.; Store the historical video assets in Cloud Storage using the Archive Storage class with Object Lifecycle Management to automate tiering.

Answer

The architect should execute the batch video rendering tasks using Cloud Run Jobs and store the historical video assets in Cloud Storage using the Archive Storage class with Object Lifecycle Management.
Selecting Cloud Run Jobs satisfies the requirement to process intermittent stateless tasks without incurring baseline compute costs, as serverless workloads scale to zero when idle. Storing long-term rarely accessed assets in Cloud Storage Archive Storage provides the lowest storage cost per gigabyte for data accessed less than once annually while maintaining standard Cloud Storage API accessibility.

Step-by-Step Solution

1
Analyze compute workload requirements for intermittent stateless batch processing
Identified serverless container execution (Cloud Run Jobs) as optimal to avoid paying for idle compute nodes.
Batch workloads that run periodically do not require dedicated Kubernetes nodes, making fully managed serverless compute the most cost-efficient choice.
2
Analyze storage access patterns for 600 TB600\text{ TB} of historical media files
Selected Cloud Storage Archive Storage tier with Object Lifecycle Management.
Data accessed less than once per year should utilize Archive Storage to achieve the lowest cost per GB/month while preserving standard API access.

Key Concept

Serverless compute right-sizing and object storage tiering for cost optimization
Question 228Question →

A software company is building a mobile gaming platform on Google Cloud. The architecture team needs to choose managed storage services for two distinct workloads:

1. User profiles and game session states requiring low-latency NoSQL document storage with high availability and a flexible schema.
2. Binary game asset files (graphics and audio) requiring object storage with automated rules to transition older files to lower-cost archival tiers.

Which TWO Google Cloud storage services should the architect recommend?

Select all that apply

Show answer & explanation

Answer: Cloud Firestore for storing document-based user profiles and game state data; Cloud Storage with Object Lifecycle Management for storing binary asset files

Answer

Cloud Firestore for user profiles and game state data, combined with Cloud Storage utilizing Object Lifecycle Management for binary game assets.
Cloud Firestore is the ideal serverless NoSQL document database for managing flexible user profiles and state documents with low latency. Cloud Storage is designed specifically for unstructured binary assets (graphics and audio) and provides native Object Lifecycle Management rules to automatically downgrade objects to colder, cost-optimized storage classes over time.

Step-by-Step Solution

1
Analyze the requirements for the first workload (User profiles and game session states).
Identified access pattern: low-latency, flexible schema, NoSQL document store.
Cloud Firestore is Google Cloud's primary managed serverless NoSQL document database.
2
Analyze the requirements for the second workload (Binary game assets with lifecycle transition).
Identified access pattern: unstructured binary object storage with automated lifecycle management.
Cloud Storage stores unstructured media files and supports Object Lifecycle Management policies.
3
Evaluate distractors against workload constraints.
Cloud Spanner is unsuitable for non-relational document data, and Transfer Appliance is an offline migration physical hardware device.
Aligning service capabilities directly with workload constraints eliminates incorrect products.

Key Concept

Selecting GCP storage services based on data structure (document NoSQL vs unstructured binary objects) and lifecycle automation features.
Question 229Question →

A global logistics and freight management company is modernizing its on-premises dispatch tracking system to Google Cloud to resolve significant architectural technical debt. The legacy system relies on microservices accessing a shared 20 TB MySQL database directly, causing tight coupling and schema migration failures. The business demands zero downtime during migration and a decoupled architecture moving forward.

Which of the following architectural actions should the cloud architect recommend to mitigate technical debt and execute the migration successfully? (Select TWO.)

Select all that apply

Show answer & explanation

Answer: Use Database Migration Service (DMS) with continuous change data capture (CDC) to sync to Cloud SQL for MySQL, and utilize the expand-and-contract pattern to decouple application deployments from schema changes.; Refactor application microservices to communicate via REST APIs and Cloud Pub/Sub asynchronous messaging instead of querying shared database tables directly.

Answer

The correct recommendations are using Database Migration Service with continuous CDC replication combined with the expand-and-contract pattern, and refactoring microservices to communicate via APIs and Cloud Pub/Sub messaging instead of accessing shared database tables directly.
To manage technical debt during cloud migration, legacy shared database dependencies must be eliminated by introducing API layers and event messaging (Cloud Pub/Sub). Additionally, migrating live data with Database Migration Service using CDC alongside an expand-and-contract schema strategy enables zero-downtime migration and safe schema evolution.

Step-by-Step Solution

1
Evaluate data migration and schema coupling strategies.
Database Migration Service (DMS) provides zero-downtime continuous replication from on-premises MySQL to Cloud SQL. Applying the expand-and-contract pattern ensures backwards-compatible schema changes during transition.
This resolves legacy schema coupling debt while maintaining zero downtime during migration.
2
Address service coupling and database access technical debt.
Decoupling services by introducing API endpoints and Cloud Pub/Sub event streams removes direct database sharing between microservices.
This establishes clear service boundaries and prevents cross-service database locks and rigid dependencies.

Key Concept

Managing Technical Debt and Decoupling Legacy Cloud Migrations
Question 230Question →

A regional retail enterprise is migrating a stateless REST API backend for its customer loyalty application to Google Cloud. The application experiences variable traffic patterns with heavy request volume during daytime business hours and virtually zero traffic overnight. The company's primary business objective is to minimize baseline infrastructure costs and operational management overhead while maintaining high availability during operational hours. Which deployment architecture best meets these business and cost requirements?

Show answer & explanation

Answer: Deploy the containerized REST API application to Cloud Run and allow it to scale down to zero instances during idle periods.

Answer

Deploying the containerized application to Cloud Run with scale-to-zero capability minimizes operational overhead and eliminates idle baseline infrastructure costs.
Deploying the stateless REST API to Cloud Run directly aligns with business goals by providing a fully managed serverless environment. Cloud Run scales automatically based on HTTP traffic, scaling down to zero instances during overnight idle periods so the enterprise incurs no compute costs when the app is not in use, while also eliminating cluster management overhead.

Step-by-Step Solution

1
Analyze workload characteristics and business requirements.
Workload is a stateless REST API with high daytime traffic and zero overnight traffic requiring minimal operational overhead and zero idle cost.
Matching compute resources to workload state and usage patterns is essential for cost optimization.
2
Evaluate Google Cloud compute options against operational and cost constraints.
Cloud Run automatically scales containerized stateless services down to zero instances, charging only during request processing time.
Serverless execution avoids paying for unutilized compute infrastructure during non-business hours.
3
Eliminate sub-optimal architectures.
GKE Standard, Compute Engine CUDs for peak capacity, and Cloud Spanner introduce fixed baseline costs and excess operational complexity.
Fixed or over-provisioned infrastructure contradicts the objective of minimizing idle costs for variable workloads.

Key Concept

Serverless compute right-sizing and scale-to-zero cost optimization for stateless workloads
Estimated Time:1m 30s
Question 231Question →

An enterprise e-commerce platform processes asynchronous product image re-indexing and media processing jobs using Compute Engine Managed Instance Groups (MIGs) fed by a Cloud Pub/Sub topic. During high-traffic marketing events, the worker nodes experience network I/O wait state spikes, resulting in processing backlogs even though average CPU utilization remains below 35%. Additionally, rapidly launching new instances during traffic surges occasionally fails due to resource availability boundaries. Which TWO architectural actions should you implement to ensure effective auto-scaling and capacity reliability? (Select TWO)

Select all that apply

Show answer & explanation

Answer: Configure the MIG autoscaler to scale based on a custom Cloud Monitoring metric tracking Pub/Sub unacknowledged message backlog per instance rather than CPU utilization.; Request regional Compute Engine quota increases for vCPUs and in-use IP addresses well before expected high-traffic events to prevent capacity allocation failures.

Answer

The correct actions are to configure the MIG autoscaler to scale based on custom metrics such as Pub/Sub unacknowledged message backlog per instance, and to proactively request regional Compute Engine quota increases for vCPUs and IP addresses prior to high-traffic events.
Scaling queue-based, I/O-bound processing requires metrics that track queue depth or backlog per instance because CPU utilization stays low during network and storage waits. Furthermore, auto-scaling relies on underlying GCP infrastructure capacity; requesting regional vCPU and networking quota increases in advance guarantees that instances can launch successfully when scale-out conditions are met.

Step-by-Step Solution

1
Identify the workload metric bottleneck
Recognize that I/O-bound asynchronous queue processing workloads keep CPU utilization artificially low while work accumulates.
Scaling on CPU utilization fails for I/O-bound tasks; tracking queue depth or custom metrics accurately reflects scaling demand.
2
Identify capacity planning requirements
Recognize that auto-scaling mechanisms cannot exceed pre-allocated GCP project resource quotas.
Proactive quota management prevents scaling execution failures when traffic surges require launching additional instances.

Key Concept

Auto-scaling metric selection for I/O-bound workloads and proactive regional quota management
Question 232Question →

A healthcare telemetry platform manages its Google Cloud infrastructure using Terraform. The cloud engineering team needs to establish a backend configuration for storing Terraform state files across dev, staging, and production environments. The strategy must prevent concurrent deployment race conditions, protect state files against accidental overwrites or corruption, enforce least-privilege administrative access, and support full operational auditability. Which configuration best meets these requirements?

Show answer & explanation

Answer: Provision a Google Cloud Storage bucket with Object Versioning enabled and uniform bucket-level access, configured as a remote backend with native state locking.

Answer

Provision a Google Cloud Storage bucket with Object Versioning enabled and uniform bucket-level access, configured as a remote backend with native state locking.
Configuring a Google Cloud Storage bucket with Object Versioning enabled, uniform bucket-level access, and native state locking satisfies all governance, durability, and concurrency requirements. Native state locking prevents concurrent pipeline executions from corrupting state, Object Versioning permits recovery from accidental overwrites or corruption, and uniform bucket-level access paired with granular IAM roles enforces least privilege.

Step-by-Step Solution

1
Identify key operational requirements for production-grade Terraform state management
Requirements include remote centralized state storage, state locking to prevent race conditions during concurrent runs, Object Versioning for disaster recovery, and fine-grained access control.
Infrastructure as Code state files store critical cloud resources mappings and require durability, concurrency control, and point-in-time recovery capabilities.
2
Evaluate Google Cloud Storage features for Terraform state backends
GCS backends natively support state locking using storage object metadata locks, while GCS Object Versioning provides automated history retention for state restoration.
Using native GCS backend capabilities fulfills high-availability and security requirements without adding external database complexity.
3
Apply Security Least-Privilege IAM Controls
Enforcing uniform bucket-level access and assigning granular predefined storage roles (such as Storage Object Admin for pipeline service accounts) ensures proper operational security.
Avoiding primitive IAM roles ensures compliance with Google Cloud security best practices.

Key Concept

Terraform Remote Backend State Management and Security Governance in Cloud Storage
Estimated Time:1m 30s
Question 233Question →

A global pharmaceutical enterprise is architecting a mission-critical platform to track international clinical trial supply chains across North America and Europe. The solution must support relational schema enforcement and strong ACID transactional consistency for writes across both continents to satisfy strict regulatory compliance. The workload requires horizontal write scaling to handle unpredictable surges in telemetry log entries without manual database sharding, while continuing to support standard ANSI SQL queries for reporting tools. Which Google Cloud storage service should the lead cloud architect select to meet these requirements?

Show answer & explanation

Answer: Cloud Spanner configured with a multi-region instance configuration across North America and Europe.

Answer

Cloud Spanner configured with a multi-region instance configuration across North America and Europe.
The correct option specifies Cloud Spanner configured across multiple regions. Cloud Spanner is Google Cloud's fully managed, enterprise-grade relational database designed specifically for workloads requiring global multi-region scale, horizontal write scaling, ANSI SQL compliance, and strong ACID transactional consistency.

Step-by-Step Solution

1
Analyze database requirements
Identified core requirements: relational schema, multi-region ACID transactions with strong consistency for writes across continents, horizontal write scaling, and ANSI SQL query support.
These constraints rule out unstructured object storage and databases incapable of cross-region synchronous write transactions.
2
Evaluate relational database options in Google Cloud
Cloud SQL only scales reads via asynchronous replicas across regions and cannot scale writes horizontally or provide multi-region synchronous ACID writes. Cloud Spanner utilizes atomic clocks (TrueTime) to provide global ACID transactions and horizontal write scaling.
Cloud Spanner is the only managed Google Cloud relational service engineered for global, multi-region active-active transactional consistency without manual database sharding.
3
Select optimal service architecture
Recommend Cloud Spanner with a multi-region deployment topology.
Satisfies all transactional consistency, scalability, and query requirements simultaneously.

Key Concept

Multi-Region Relational Storage Selection and Cloud Spanner vs Cloud SQL Architecture
Question 234Question →

A global financial technology enterprise is architecting its core payment processing engine on Google Cloud. The workload requires strong transactional consistency across multiple geographic regions with a strict Recovery Point Objective (RPO) of zero and a Recovery Time Objective (RTO) under 5 seconds in the event of a regional outage. Additionally, the stateless API microservices tier running on Compute Engine Managed Instance Groups (MIGs) behind an External HTTP(S) Load Balancer must prevent cascading instance recreations and service degradation during temporary database latency spikes. Which deployment architecture fulfills these technical and high availability requirements?

Show answer & explanation

Answer: Deploy a multi-region Cloud Spanner configuration for the database tier, and configure the MIG load balancer health checks to query a lightweight local application endpoint that verifies process health without invoking downstream database queries.

Answer

Deploying a multi-region Cloud Spanner database paired with Managed Instance Groups configured with shallow local health checks satisfies the zero RPO multi-region availability requirement while preventing cascading failure modes during traffic spikes.
Cloud Spanner multi-region configurations deliver multi-region high availability with synchronous replication, achieving zero RPO and automatic cross-region failover (RTO < 5 seconds). Coupling this with shallow HTTP health checks on the API compute tier isolates web server operational status from transient backend database latency, avoiding cascading failures.

Step-by-Step Solution

1
Evaluate the database tier requirements for RPO = 0 and sub-5-second RTO across multiple regions.
Cloud Spanner with a multi-region instance deployment uses Paxos consensus and TrueTime to provide synchronous cross-region write replication, achieving RPO = 0 and automatic failover.
Cloud SQL regional HA only protects against single-zone failures within a single region. Cross-region Cloud SQL replicas use asynchronous replication which cannot guarantee RPO = 0.
2
Evaluate the compute tier high availability and health check configuration for stateless MIG instances.
Load balancer health checks must assess the local HTTP/gRPC responsiveness of the compute instance itself (shallow check) rather than querying backend dependencies.
Deep health checks that test backend databases cause false-positive failures during transient DB latency spikes, causing the load balancer to drain traffic and terminate healthy compute nodes, exacerbating the outage.

Key Concept

Multi-Region High Availability and Health Check Isolation
Estimated Time:2m 0s
Question 235Question →

An online gaming platform is designing a hybrid network architecture to connect its on-premises match-making datacenter to two separate Google Cloud VPC networks: `vpc-game-prod` (hosting game servers) and `vpc-analytics` (hosting telemetry pipelines). The architecture must meet the following technical requirements:
- Baseline hybrid bandwidth demand is 6 Gbps continuously with high SLA requirements (99.99% availability).
- Both VPC networks are connected to each other using VPC Network Peering.
- On-premises hosts must reach both `vpc-game-prod` and `vpc-analytics` directly without attempting to route traffic transitively through VPC Peering.

Which TWO architectural decisions must be implemented to fulfill these requirements? (Select TWO)

Select all that apply

Show answer & explanation

Answer: Provision Dedicated Interconnect connections with separate VLAN attachments configured on Cloud Routers in both `vpc-game-prod` and `vpc-analytics`.; Provision four Dedicated Interconnect circuits deployed across two separate colocation facilities (metros) with redundant Cloud Routers in each region.

Answer

The correct architecture requires provisioning Dedicated Interconnect connections directly into both `vpc-game-prod` and `vpc-analytics` to handle non-transitive VPC Peering limits, and deploying four Dedicated Interconnect circuits across two metro locations to meet the 99.99% availability SLA.
To achieve 99.99% availability for hybrid connectivity with a 6 Gbps baseline requirement, GCP guidelines mandate deploying Dedicated Interconnect across two distinct metro locations with four total connections. Additionally, because VPC Network Peering is strictly non-transitive, hybrid connectivity must be provisioned with Cloud Routers and VLAN attachments in both VPC networks so on-premises traffic can directly access each VPC.

Step-by-Step Solution

1
Evaluate throughput and SLA constraints
A 6 Gbps continuous throughput requirement with 99.99% availability mandates Dedicated Interconnect with 4 connections across 2 metro locations.
Dedicated Interconnect provides 10 Gbps or 100 Gbps physical circuits. Achieving a 99.99% SLA requires 4 VLAN attachments across 2 distinct metro colocation facilities.
2
Evaluate VPC Peering routing capabilities
VPC Peering does not support transitive routing.
On-premises resources cannot reach `vpc-analytics` by routing through `vpc-game-prod` over VPC Peering. Therefore, hybrid interconnect connections must be terminated into both VPC networks.

Key Concept

Designing 99.99% SLA Dedicated Interconnect topologies and addressing VPC Network Peering non-transitivity.
Question 236Question →

An enterprise architecture team is implementing an automated Infrastructure as Code (IaC) pipeline using Cloud Build and Terraform to provision Google Cloud environments across multiple projects. The pipeline must securely handle state management, prevent state file corruption during concurrent executions, and enforce the principle of least privilege for automated deployments. Which TWO actions should the team implement to satisfy these requirements?

Select all that apply

Show answer & explanation

Answer: Configure a Google Cloud Storage bucket backend for Terraform state with Object Versioning enabled to support native state locking and state restoration.; Grant the Cloud Build service account specific predefined roles required for the target resources, along with the Service Account User role (roles/iam.serviceAccountUser) on designated runtime service accounts.

Answer

Configure a Cloud Storage bucket backend with Object Versioning for state locking, and grant the Cloud Build service account specific predefined resource roles with roles/iam.serviceAccountUser on target service accounts.
Centralizing Terraform state storage in Google Cloud Storage with versioning ensures state consistency, encryption, and locking mechanism support across concurrent Cloud Build runs. Furthermore, restricting the Cloud Build service account to fine-grained predefined roles and granting roles/iam.serviceAccountUser specifically on required service accounts enforces least privilege while enabling successful compute resource attachment.

Step-by-Step Solution

1
Establish secure and robust Terraform remote state storage.
Configuring a Cloud Storage backend with Object Versioning enables centralized remote state with native locking capabilities to prevent race conditions during automated deployments.
Remote state in Cloud Storage ensures consistency, encryption at rest, access control via IAM, and concurrency management.
2
Configure CI/CD IAM permissions following the principle of least privilege.
Assign fine-grained predefined roles for provisioned infrastructure and grant roles/iam.serviceAccountUser on specific compute service accounts.
This grants the Cloud Build service account only the privileges necessary to create resources and attach service accounts, avoiding excessive administrative access.

Key Concept

Terraform remote state management with GCS and least-privilege IAM configuration for CI/CD automation pipelines.
Estimated Time:2m 0s
Question 237Question →

An enterprise logistics provider is architecting a mission-critical fleet management service on Google Cloud within a single GCP region. The workload consists of a lightweight, stateless HTTP microservice that experiences fluctuating request volumes, and a relational database backend requiring multi-zone high availability with automatic failover to satisfy a strict regional Uptime SLA. The backend database does not require global distribution or horizontal write scaling. Which architectural design meets these technical and high availability requirements while minimizing operational overhead and unnecessary infrastructure costs?

Show answer & explanation

Answer: Deploy the microservice on Cloud Run integrated via a Serverless VPC Access connector to a regional Cloud SQL instance configured for High Availability (HA) across multiple zones.

Answer

Deploy the microservice on Cloud Run integrated via a Serverless VPC Access connector to a regional Cloud SQL instance configured for High Availability (HA) across multiple zones.
Combining Cloud Run with a regional Cloud SQL HA deployment directly satisfies the requirement for multi-zone high availability, automated failover, and scaling under fluctuating traffic while keeping operational overhead and costs minimal.

Step-by-Step Solution

1
Analyze Compute Requirements
Identify that Cloud Run is optimal for stateless HTTP microservices with variable traffic because it automatically scales down to zero and eliminates GKE cluster management overhead.
Choosing serverless compute reduces operational toil while ensuring high availability across zones within the region.
2
Analyze Database Requirements
Select Cloud SQL with High Availability (HA) enabled, which uses synchronous storage replication between primary and standby instances across two zones in the same region.
Cloud SQL HA fulfills regional zero-data-loss RPO and fast RTO automatic failover requirements without needing Cloud Spanner's global distributed scale.
3
Configure Secure Internal Connectivity
Connect Cloud Run to Cloud SQL using Serverless VPC Access connector to route traffic via private IP.
Serverless VPC Access enables private connectivity between Google serverless services and internal VPC resources.

Key Concept

Designing High Availability for Regional Web Application Tiers
Estimated Time:1m 30s
Question 238Question →

A global healthcare analytics platform is migrating its patient portal to Google Cloud. The architecture consists of a stateless web tier processing patient query requests and a database backend operating within a single Google Cloud region. The design must accommodate 12 Gbps of low-latency, SLA-backed hybrid connectivity between the on-premises electronic health record (EHR) database and the VPC network. Additionally, the web tier must maintain an availability target of 99.99% against zonal outages and prevent health monitoring probes from causing cascading service disruptions if backend database latency spikes. Which deployment architecture meets these technical requirements?

Show answer & explanation

Answer: Deploy the web application on Cloud Run across multiple zones behind an External HTTP(S) Load Balancer using shallow endpoint health checks (/healthz), and configure Dedicated Interconnect with redundant VLAN attachments across two metropolitan edge locations for hybrid connectivity.

Answer

Deploy the web application on Cloud Run across multiple zones behind an External HTTP(S) Load Balancer using shallow endpoint health checks (/healthz), and configure Dedicated Interconnect with redundant VLAN attachments across two metropolitan edge locations for hybrid connectivity.
Dedicated Interconnect with dual attachments across two edge facilities provides the required 12 Gbps capacity with 99.99% SLA. Coupling this with Cloud Run behind Cloud Load Balancing and shallow health checks (/healthz) ensures high availability across zones without risking cascading failures from database query latency.

Step-by-Step Solution

1
Evaluate hybrid network throughput requirements
12 Gbps throughput exceeds the standard 3 Gbps limit per HA VPN tunnel, requiring GCP Dedicated Interconnect across multiple interconnect locations to provide high throughput and SLA guarantees.
Dedicated Interconnect is designed for enterprise workloads requiring >10 Gbps dedicated throughput and low latency connectivity.
2
Evaluate health check strategy for high availability
Load balancer health checks must probe a shallow HTTP endpoint on the application container rather than deep downstream dependencies like the database.
Deep health check endpoints cause false-positive failures and cascading reboots across compute tiers whenever downstream databases experience transient load.
3
Select compute and regional HA tier
Cloud Run or multi-zone MIGs behind Cloud Load Balancing satisfy stateless application scaling across availability zones.
Fully managed multi-zone serverless compute combined with dedicated hybrid connections meets the 99.99% availability target cleanly.

Key Concept

Hybrid Interconnect Topology & Load Balancer Health Check Isolation
Question 239Question →

A Software-as-a-Service (SaaS) enterprise is designing the core architecture for its customer analytics portal on Google Cloud. The solution requires multi-regional high availability with a Recovery Time Objective (RTO) near zero and a Recovery Point Objective (RPO) under 5 seconds. The workload consists of a transactional relational database that demands strong global consistency across regions, alongside a stateless HTTP microservices API layer with fluctuating request volumes. Which TWO architectural decisions should you make to satisfy these technical requirements and high availability targets?

Select all that apply

Show answer & explanation

Answer: Provision a multi-region Cloud Spanner instance for the transactional relational database layer.; Deploy the stateless microservices on Cloud Run across multiple regions behind a Global External HTTP(S) Load Balancer.

Answer

To meet strict multi-region high availability, near-zero RTO, and sub-5-second RPO with strong relational consistency, the architecture requires Cloud Spanner in a multi-region configuration for data persistence and Cloud Run deployed across multiple regions behind a Global External HTTP(S) Load Balancer for the compute tier.
Cloud Spanner in a multi-region configuration delivers built-in synchronous multi-region replication and strong consistency, directly fulfilling the strict RPO and RPO requirements. Deploying stateless HTTP microservices on Cloud Run across multiple regions behind a Global External HTTP(S) Load Balancer provides automatic scaling, low latency traffic distribution, and seamless regional failover without operational cluster management.

Step-by-Step Solution

1
Evaluate data layer requirements for multi-region high availability and consistency.
Cloud Spanner multi-region provides synchronous replication across regions, ensuring strong consistency and sub-second RPO.
Standard relational services like Cloud SQL rely on asynchronous cross-region replicas which cannot guarantee synchronous multi-region consistency or zero RPO.
2
Evaluate compute layer requirements for stateless microservices under variable demand.
Cloud Run deployed across multiple regions integrated with a Global External HTTP(S) Load Balancer automatically scales compute resources based on incoming traffic while providing multi-region failover.
Using serverless Cloud Run eliminates cluster management overhead compared to GKE, while the global load balancer routes traffic to the nearest healthy region.

Key Concept

Designing multi-region highly available architectures using Cloud Spanner for globally consistent relational data and Cloud Run with Global Load Balancing for stateless compute.
Question 240Question →

A regional e-commerce platform is re-architecting its containerized order reservation backend on Google Cloud. The application consists of stateless HTTP services handling unpredictable traffic spikes during promotional events. The solution must run within a single GCP region while ensuring high availability across multiple zones, automatic scaling from zero, and minimal cluster management overhead. The underlying relational database requires ACID transactional integrity within the region and automatic standby failover. Which architecture meets these technical and high availability requirements with the lowest operational complexity?

Show answer & explanation

Answer: Deploy the stateless microservices on Cloud Run connected via Serverless VPC Access to a regional Cloud SQL instance configured with High Availability (HA).

Answer

Deploying the stateless microservices on Cloud Run coupled with Serverless VPC Access to a regional Cloud SQL for PostgreSQL/MySQL HA instance satisfies high availability, auto-scaling, and single-region transactional requirements while minimizing operational overhead.
The combination of Cloud Run and Cloud SQL in High Availability (HA) mode perfectly satisfies all technical requirements. Cloud Run provides serverless container execution that scales seamlessly with traffic spikes while eliminating cluster maintenance overhead. Cloud SQL HA uses synchronous replication across two zones within the region to deliver automatic failover and high availability for relational workloads.

Step-by-Step Solution

1
Evaluate compute tier requirements against operational overhead constraints.
Cloud Run handles stateless HTTP containers with automatic scaling (including scale-to-zero) and requires zero infrastructure management, making it superior to GKE for low operational overhead.
The requirement explicitly asks to minimize cluster management and operational complexity.
2
Evaluate database tier requirements for single-region ACID transactional compliance and high availability.
Cloud SQL configured in High Availability (HA) mode provides a primary instance in one zone and a synchronous standby instance in a second zone within the same region.
Cloud SQL HA meets regional RPO=0 synchronous replication and fast failover needs without the cost and operational complexity of globally distributed databases like Cloud Spanner.
3
Determine secure private networking between serverless compute and database resources.
Serverless VPC Access connector allows Cloud Run instances to route traffic privately into the VPC network to reach internal Cloud SQL IP addresses.
Direct private IP communication protects database endpoints without exposing public IP addresses.

Key Concept

Designing serverless high availability architectures using Cloud Run and Cloud SQL HA
PreviousPage 12 / 80Next
All practice questions — Google Cloud Professional Cloud Architect | Examkin