Google Cloud Professional Cloud Architect - Fact Sheet¶

Quick Reference¶

Exam Code: Professional Cloud Architect Duration: 120 minutes (2 hours) Questions: 50-60 questions Passing Score: ~70% (not officially published) Cost: $200 USD Validity: 2 years Difficulty: ⭐⭐⭐⭐⭐ (Most challenging GCP certification) Prerequisites: Recommended 3+ years of industry experience, including 1+ year designing GCP solutions

Exam Domains¶

Core Architecture Principles¶

Well-Architected Framework¶

📖 Google Cloud Architecture Framework - Complete architecture framework

Five Pillars: 1. Operational Excellence - Efficient operations and monitoring 2. Security, Privacy, and Compliance - Protection and governance 3. Reliability - Availability and fault tolerance 4. Cost Optimization - Resource efficiency and financial governance 5. Performance Optimization - Efficient resource use and scaling

Key Resources: - 📖 Architecture Center - Reference architectures and diagrams - 📖 Best Practices - Enterprise design patterns - 📖 Cloud Architecture Patterns - System design patterns - 📖 Decision Trees - Architecture decision guides

Compute Services - Architecture Deep Dive¶

Compute Engine¶

Strategic Use Cases: - Legacy application migrations (lift-and-shift) - Custom operating systems or kernels - GPU/TPU workloads requiring specific configurations - Per-second billing with sustained use discounts - 📖 Compute Engine Overview - Architecture guide - 📖 Machine Families - Choosing the right machine type - 📖 Instance Templates - Configuration management - 📖 Managed Instance Groups - Auto-scaling and load balancing - 📖 Sole-Tenant Nodes - Dedicated hardware for compliance

Advanced Features: - Custom Machine Types - Precise resource allocation - Preemptible VMs - Up to 80% cost savings for fault-tolerant workloads - Spot VMs - More flexible than preemptible with longer runtime - Committed Use Discounts - 1 or 3 year commitments for 57% savings - Shielded VMs - Secure boot, vTPM, integrity monitoring - 📖 Live Migration - Zero-downtime maintenance - 📖 Persistent Disk Performance - Storage optimization

Google Kubernetes Engine (GKE)¶

Enterprise Architecture: - Autopilot - Fully managed, pay-per-pod, optimal configurations - Standard - Node-level control, custom configurations - GKE Enterprise - Multi-cluster management with Anthos - 📖 GKE Architecture - Complete overview - 📖 Cluster Architecture - Control plane and nodes - 📖 Multi-Cluster Ingress - Global load balancing - 📖 Binary Authorization - Deploy-time security controls - 📖 GKE Networking - VPC-native clusters

Advanced Patterns: - Workload Identity - Pod-to-GCP service authentication without keys - Config Connector - Manage GCP resources via Kubernetes - Node Auto-Provisioning - Automatic node pool creation - Vertical Pod Autoscaling - Right-size container resources - Multi-cluster Services - Service discovery across clusters - 📖 GKE Best Practices - Production guidelines - 📖 Security Hardening - Security best practices

Cloud Run¶

Serverless Container Architecture: - Fully managed compute platform for containers - Scales to zero, pay only for actual usage - Auto-scales based on concurrent requests - 📖 Cloud Run Documentation - Complete guide - 📖 Service Identity - IAM integration - 📖 VPC Connectivity - Direct VPC egress - 📖 Cloud Run Jobs - Batch workloads

Architecture Decisions: - vs App Engine - More flexibility, custom containers - vs GKE - Simpler operations, less control - vs Cloud Functions - Longer execution time, more memory

App Engine¶

Platform as a Service: - Standard Environment - Rapid scaling, sandbox runtime, free tier - Flexible Environment - Docker containers, SSH access, custom runtimes - 📖 App Engine Architecture - Environment comparison - 📖 Scaling Types - Automatic, basic, manual scaling - 📖 Traffic Splitting - A/B testing and canary - 📖 Migration to Standard 2^nd Gen - Modern runtimes

Cloud Functions¶

Event-Driven Serverless: - 2^nd Generation - Built on Cloud Run, better performance, longer execution - Event sources: Cloud Storage, Pub/Sub, HTTP, Firestore, Firebase - 📖 Cloud Functions Architecture - Concepts and patterns - 📖 Eventarc Integration - Unified eventing - 📖 Security Best Practices - Function security

Storage Architecture¶

Cloud Storage¶

Global Object Storage: - Storage Classes Decision Matrix: - Standard - Hot data, frequent access - Nearline - < 1/month access, 30-day minimum - Coldline - < 1/quarter access, 90-day minimum - Archive - < 1/year access, 365-day minimum - 📖 Storage Classes - Detailed comparison - 📖 Object Lifecycle Management - Automated tiering - 📖 Bucket Locations - Regional, dual-region, multi-region - 📖 Object Versioning - Version control - 📖 Retention Policies - Compliance and data governance - 📖 Customer-Managed Encryption Keys - CMEK integration

Advanced Features: - Requester Pays - Transfer costs to downloader - Object Holds - Legal and temporary holds - Dual-region - 99.95% SLA with geo-redundancy - Turbo Replication - < 15 minute RPO between regions - 📖 Performance Optimization - Request rate best practices

Persistent Disks and Filestore¶

Block and File Storage: - Persistent Disk Types: - Standard (pd-standard) - HDD, sequential workloads - Balanced (pd-balanced) - SSD, balanced price/performance - SSD (pd-ssd) - High-performance SSD - Extreme (pd-extreme) - Highest performance, configurable IOPS - 📖 Disk Types Comparison - Performance and pricing - 📖 Regional Persistent Disks - Synchronous replication - 📖 Disk Snapshots - Incremental backups - 📖 Filestore - Managed NFS file server - 📖 Filestore Instances - Basic, High Scale, Enterprise tiers

Database Architecture¶

Cloud SQL¶

Managed Relational Databases: - Supports MySQL, PostgreSQL, SQL Server - High Availability Configuration: - Regional HA with synchronous replication - Automatic failover (typically 60-120 seconds) - Read replicas for read scaling - 📖 Cloud SQL Overview - Architecture guide - 📖 High Availability - HA architecture - 📖 Replication - Read replicas and cross-region - 📖 Backup and Recovery - Backup strategies - 📖 Connection Options - Cloud SQL Proxy, Private IP

Cloud Spanner¶

Globally Distributed Relational Database: - Horizontally scalable, strongly consistent - Global transactions with 99.999% availability SLA - Multi-region configurations for HA and low latency - 📖 Cloud Spanner Overview - Architecture concepts - 📖 Replication - Multi-region replication - 📖 Schema Design Best Practices - Performance optimization - 📖 Instance Configurations - Regional and multi-regional - 📖 Choosing Between Cloud SQL and Spanner - Decision guide

Firestore and Datastore¶

NoSQL Document Databases: - Firestore - Next-generation, real-time synchronization - Datastore Mode - Server-side applications - Automatic multi-region replication - Strong consistency for entity group queries - 📖 Firestore Overview - Complete guide - 📖 Choosing Between Firestore Modes - Native vs Datastore mode - 📖 Data Modeling - Document structure - 📖 Best Practices - Performance and cost optimization

Bigtable¶

Wide-Column NoSQL: - Petabyte-scale, sub-10ms latency - Ideal for time-series, IoT, financial services - HBase API compatible - 📖 Bigtable Overview - Architecture and use cases - 📖 Schema Design - Row key design critical - 📖 Replication - Multi-cluster replication - 📖 Performance Tuning - Optimization guide - 📖 Choosing Between Bigtable and Other Databases - Decision tree

BigQuery¶

Serverless Data Warehouse: - Petabyte-scale analytics with standard SQL - Separation of compute and storage - Automatic optimization and indexing - 📖 BigQuery Architecture - Complete guide - 📖 Partitioning and Clustering - Query optimization - 📖 BigQuery BI Engine - In-memory analytics - 📖 BigQuery ML - Machine learning in SQL - 📖 Cost Optimization - Query cost management - 📖 Data Transfer Service - Scheduled data imports

Memorystore¶

Managed In-Memory Data Store: - Memorystore for Redis - Advanced features, persistence - Memorystore for Memcached - Simple caching - 📖 Memorystore for Redis - Redis architecture - 📖 High Availability - Standard vs basic tier

Networking Architecture¶

Virtual Private Cloud (VPC)¶

Network Foundation: - Global resource spanning all regions - Subnets are regional (not zonal) - No inter-region bandwidth charges within VPC - 📖 VPC Overview - Fundamental concepts - 📖 VPC Network Design - Best practices - 📖 Subnet Configuration - IP addressing - 📖 Firewall Rules - Traffic control - 📖 Routes - Routing configuration - 📖 Alias IP Ranges - Container and service IPs

Advanced VPC Patterns: - Shared VPC - Cross-project networking - VPC Peering - Private connectivity between VPCs - VPC Service Controls - Security perimeters - Private Google Access - Access Google APIs privately - Private Service Connect - Private access to managed services - 📖 Shared VPC - Multi-project architecture - 📖 VPC Peering - VPC interconnection - 📖 VPC Service Controls - Security perimeter - 📖 Private Google Access - API access without internet - 📖 Private Service Connect - Private service access

Cloud Load Balancing¶

Global Load Balancing: - Single anycast IP serving globally - Automatic multi-region failover - Layer 4 (TCP/UDP) and Layer 7 (HTTP/S) options - 📖 Load Balancing Overview - Architecture guide - 📖 Choosing a Load Balancer - Decision tree - 📖 External HTTP(S) Load Balancer - Global Layer 7 - 📖 Internal HTTP(S) Load Balancer - Regional Layer 7 - 📖 Network Load Balancer - Regional Layer 4 - 📖 SSL Policies - TLS configuration - 📖 Cloud Armor - DDoS protection and WAF

Cloud CDN¶

Content Delivery Network: - Global edge network with Google's infrastructure - Cache-to-cache filling reduces origin load - Integration with Cloud Load Balancing - 📖 Cloud CDN Overview - Architecture - 📖 Caching Best Practices - Performance optimization - 📖 Cache Keys and Modes - Cache control - 📖 Signed URLs and Cookies - Content access control

Hybrid Connectivity¶

Connecting to Google Cloud: - Cloud VPN - IPsec, up to 3 Gbps per tunnel with HA VPN - Cloud Interconnect - Dedicated physical connections - Dedicated - 10 or 100 Gbps direct connection - Partner - 50 Mbps to 50 Gbps via partner - 📖 Hybrid Connectivity Overview - Options comparison - 📖 Cloud VPN - VPN architecture - 📖 HA VPN - High availability VPN - 📖 Cloud Interconnect - Dedicated connectivity - 📖 Partner Interconnect - Service provider connectivity - 📖 Cloud Router - Dynamic BGP routing

Network Intelligence Center: - 📖 Network Topology - Visualization - 📖 Connectivity Tests - Troubleshooting - 📖 Performance Dashboard - Monitoring

Security and Identity¶

Identity and Access Management (IAM)¶

Principle of Least Privilege: - Who - Google Account, Service Account, Google Group, Cloud Identity domain - What - Resources (projects, folders, organization) - How - Roles (primitive, predefined, custom) - 📖 IAM Overview - Core concepts - 📖 IAM Roles - Role types and hierarchy - 📖 Custom Roles - Role creation - 📖 IAM Conditions - Conditional access - 📖 IAM Best Practices - Security guidelines - 📖 Policy Intelligence - Policy analysis tools

Service Accounts¶

Application Identity: - Default compute service account (not recommended for production) - User-managed service accounts (recommended) - Short-lived credentials via Workload Identity Federation - 📖 Service Accounts - Complete guide - 📖 Best Practices - Security patterns - 📖 Workload Identity - External identity federation - 📖 Service Account Impersonation - Delegation patterns

Encryption¶

Data Protection: - Encryption at rest - Default with Google-managed keys - Customer-Managed Encryption Keys (CMEK) - Cloud KMS integration - Customer-Supplied Encryption Keys (CSEK) - Customer-provided keys - 📖 Encryption at Rest - Default encryption - 📖 Cloud KMS - Key management service - 📖 CMEK - Customer-managed keys - 📖 Cloud HSM - Hardware security modules - 📖 Secret Manager - Secrets storage

Security Command Center¶

Centralized Security Management: - Asset discovery and inventory - Vulnerability scanning - Threat detection - Compliance monitoring - 📖 Security Command Center - Overview - 📖 Asset Discovery - Inventory management - 📖 Finding Types - Security findings

Identity-Aware Proxy (IAP)¶

Application-Level Access Control: - Zero-trust access to applications - No VPN required - Context-aware access controls - 📖 Identity-Aware Proxy - Architecture guide - 📖 IAP TCP Forwarding - SSH/RDP access

Operations and Observability¶

Cloud Monitoring (Operations Suite)¶

Monitoring and Alerting: - Infrastructure and application metrics - Custom metrics from applications - Uptime checks and alerting - 📖 Cloud Monitoring - Complete guide - 📖 Metrics - Available metrics - 📖 Alerting - Alert policies - 📖 Dashboards - Visualization - 📖 Uptime Checks - Availability monitoring

Cloud Logging (Operations Suite)¶

Centralized Log Management: - All GCP service logs automatically collected - Log sinks to BigQuery, Cloud Storage, Pub/Sub - Log-based metrics for custom monitoring - 📖 Cloud Logging - Architecture overview - 📖 Logs Router - Log routing - 📖 Log Sinks - Export destinations - 📖 Logs Explorer - Query interface - 📖 Audit Logs - Compliance logging

Cloud Trace and Profiler¶

Application Performance: - Cloud Trace - Distributed tracing for latency analysis - Cloud Profiler - Continuous CPU and memory profiling - Cloud Debugger - Production debugging without stopping - 📖 Cloud Trace - Latency tracking - 📖 Cloud Profiler - Performance profiling - 📖 Error Reporting - Error aggregation

Migration Strategies¶

Migration Framework (5 Rs)¶

Migration Patterns: 1. Rehost - Lift-and-shift to Compute Engine 2. Replatform - Minor optimizations (e.g., Cloud SQL instead of self-managed) 3. Refactor - Re-architect for cloud-native (serverless, microservices) 4. Retire - Decommission unnecessary systems 5. Retain - Keep on-premises temporarily

Migration Tools: - Migrate for Compute Engine - VM migration from on-prem or other clouds - Database Migration Service - Continuous replication for minimal downtime - Transfer Appliance - Physical data transfer for large datasets - Storage Transfer Service - Online data transfer - 📖 Migration to Google Cloud - Migration framework - 📖 Migrate for Compute Engine - VM migration - 📖 Database Migration Service - Database migration - 📖 Transfer Appliance - Offline data transfer - 📖 Storage Transfer Service - Online data transfer

Cost Optimization¶

Compute Cost Optimization¶

Cost-Effective Compute: - Committed Use Discounts - 1 or 3 year, up to 57% savings - Sustained Use Discounts - Automatic discounts for running instances - Preemptible VMs - Up to 80% savings, 24-hour max - Spot VMs - More flexible preemptible alternative - Custom Machine Types - Right-size CPU and memory - 📖 Pricing Overview - GCP pricing model - 📖 Committed Use Discounts - Long-term commitments - 📖 Resource-Based Pricing - Per-second billing - 📖 Pricing Calculator - Cost estimation

Storage Cost Optimization¶

Storage Classes and Lifecycle: - Autoclass for Cloud Storage buckets - Object lifecycle management - Nearline/Coldline/Archive for infrequent access - Regional vs multi-region trade-offs - 📖 Storage Cost Optimization - Best practices

Cost Management Tools¶

Visibility and Control: - Cost Management - Budgets, alerts, reports - Recommender - AI-powered cost and performance recommendations - Active Assist - Proactive optimization recommendations - 📖 Cost Management - Cost visibility - 📖 Billing Reports - Cost analysis - 📖 Budgets and Alerts - Cost controls - 📖 Recommender - Optimization recommendations

High Availability and Disaster Recovery¶

Availability Patterns¶

Multi-Zonal and Multi-Regional: - Zonal Resources - Compute Engine instances, persistent disks - Regional Resources - Cloud SQL, regional MIGs, subnets - Multi-Regional Resources - Cloud Storage, Spanner - Global Resources - VPC networks, load balancers, Cloud CDN

Disaster Recovery Strategies¶

RTO and RPO Trade-offs:

DR Best Practices: - Regular backup testing and validation - Automated failover procedures - Multi-region replication for critical data - Documentation and runbooks - Periodic DR drills

Compliance and Governance¶

Resource Hierarchy¶

Organization Structure:

Organization
├── Folder (Business Unit)
│   ├── Folder (Environment)
│   │   ├── Project (Application)
│   │   └── Project (Application)

• 📖 Resource Hierarchy - Organization structure
• 📖 Organization Policies - Centralized constraints
• 📖 Project Best Practices - Project design

Compliance and Certifications¶

Regulatory Compliance: - ISO 27001, SOC ⅔, PCI-DSS, HIPAA, GDPR - Regional data residency controls - Compliance reports and certifications - 📖 Compliance Offerings - Certifications and reports - 📖 Data Residency - Location controls

Common Architecture Patterns¶

Pattern 1: Three-Tier Web Application¶

Architecture: - Presentation Tier - Cloud Load Balancer + Cloud CDN - Application Tier - Managed Instance Group (auto-scaling) - Data Tier - Cloud SQL with read replicas

Key Services: - Global HTTP(S) Load Balancer - Cloud CDN for static assets - Cloud Armor for DDoS protection - VPC with private subnets - Cloud SQL with HA configuration

Pattern 2: Microservices on GKE¶

Architecture: - GKE Autopilot or Standard cluster - Workload Identity for service authentication - Cloud SQL Proxy for database access - Anthos Service Mesh for observability

Key Services: - GKE with VPC-native networking - Cloud Build for CI/CD - Artifact Registry for container images - Cloud Monitoring for observability

Pattern 3: Real-Time Data Pipeline¶

Architecture: - Ingestion - Pub/Sub for event streaming - Processing - Dataflow for stream processing - Storage - BigQuery for analytics - Visualization - Looker or Data Studio

Key Services: - Pub/Sub for decoupling - Dataflow for ETL - BigQuery for warehousing - Dataproc for Hadoop/Spark workloads

Pattern 4: Serverless Application¶

Architecture: - Cloud Run or Cloud Functions for compute - Firestore for NoSQL database - Cloud Storage for object storage - API Gateway for API management

Key Services: - Cloud Run with automatic scaling - Eventarc for event routing - Secret Manager for credentials - Cloud CDN for caching

Pattern 5: Hybrid Cloud with Anthos¶

Architecture: - GKE clusters on GCP and on-premises - Anthos Config Management for policy - Anthos Service Mesh for networking - Cloud Interconnect for connectivity

Key Services: - Anthos GKE on-prem - Cloud VPN or Interconnect - Binary Authorization - Cloud Operations suite

Data Engineering Architecture¶

Batch Processing¶

MapReduce and Spark: - Dataproc - Managed Hadoop and Spark - Dataflow - Serverless Apache Beam pipelines - 📖 Dataproc - Managed big data - 📖 Dataflow - Stream and batch processing

Stream Processing¶

Real-Time Analytics: - Pub/Sub - Global message queue - Dataflow - Stream processing - Bigtable - High-throughput writes - 📖 Pub/Sub - Messaging architecture - 📖 Streaming Analytics - Reference architecture

Data Warehousing¶

BigQuery Architecture: - Columnar storage for analytics - Automatic query optimization - Federated queries across sources - 📖 BigQuery Best Practices - Performance guide

Machine Learning and AI¶

AI Platform and Vertex AI¶

ML Workflow: - Vertex AI - Unified ML platform - Training with custom or pre-built containers - Model deployment and serving - 📖 Vertex AI - ML platform - 📖 AutoML - No-code ML

Pre-Trained APIs¶

AI Services: - Vision API, Natural Language API, Translation API - Speech-to-Text, Text-to-Speech - 📖 Cloud AI Products - AI/ML services

DevOps and CI/CD¶

Cloud Build¶

Continuous Integration: - Serverless build service - Container and non-container builds - Integration with GitHub, Bitbucket, Cloud Source Repositories - 📖 Cloud Build - CI/CD platform

Infrastructure as Code¶

Deployment Automation: - Deployment Manager - Native GCP IaC - Terraform - Multi-cloud IaC (popular choice) - Config Connector - Kubernetes-native GCP resource management - 📖 Deployment Manager - GCP native IaC - 📖 Terraform on GCP - Terraform integration

Exam Scenarios and Solutions¶

Scenario 1: High-Traffic Web Application¶

Requirements: Handle millions of requests, global users, cost-effective

Solution: - Global HTTP(S) Load Balancer with Cloud CDN - Regional Managed Instance Groups with autoscaling - Cloud SQL with read replicas or Cloud Spanner - Cloud Storage for static assets - Cloud Armor for security

Key Decision: Cloud Spanner if multi-region writes needed, else Cloud SQL with cross-region read replicas

Scenario 2: Lift-and-Shift Migration¶

Requirements: Migrate 100+ VMs from on-premises quickly

Solution: - Migrate for Compute Engine - Phased migration approach - Cloud VPN or Interconnect for connectivity - Use Committed Use Discounts for cost savings

Key Decision: Start with non-critical workloads, validate, then migrate production

Scenario 3: Real-Time Analytics¶

Requirements: Ingest millions of events per second, real-time dashboards

Solution: - Pub/Sub for ingestion - Dataflow for stream processing - BigQuery for data warehouse - Bigtable for operational queries - Looker or Data Studio for visualization

Key Decision: BigQuery for ad-hoc analytics, Bigtable for operational low-latency queries

Scenario 4: Hybrid Cloud¶

Requirements: Keep sensitive data on-premises, use GCP for compute

Solution: - Cloud Interconnect (dedicated) for consistent performance - Shared VPC for network segregation - Private Google Access for API calls - Cloud SQL with private IP - Anthos if running Kubernetes workloads

Key Decision: Dedicated Interconnect vs Partner Interconnect based on bandwidth needs

Scenario 5: Disaster Recovery¶

Requirements: RPO < 1 hour, RTO < 4 hours

Solution: - Pilot Light strategy - Cloud SQL automated backups - Persistent disk snapshots to Cloud Storage - Infrastructure as Code for rapid deployment - Runbooks in Cloud Storage

Key Decision: Warm Standby if stricter RTO/RPO needed

Scenario 6: Multi-Tenant SaaS¶

Requirements: Isolate customer data, cost attribution

Solution: - Separate projects per customer (strong isolation) - OR Shared infrastructure with tagging (cost-effective) - Folder hierarchy for organization - Labels for cost allocation - VPC Service Controls for data exfiltration protection

Key Decision: Isolation level vs cost trade-off

Exam Tips and Strategy¶

Keywords to Watch¶

Question Patterns: - "Cost-effective" → Committed use, preemptible, autoscaling, Cloud Functions/Run - "Minimize operational overhead" → Managed services, serverless, GKE Autopilot - "High availability" → Multi-zone, multi-region, load balancing - "Low latency" → Memorystore, CDN, proximity to users, Bigtable - "Compliance/regulatory" → VPC Service Controls, organization policies, audit logs - "Real-time" → Pub/Sub, Dataflow, Bigtable - "Big data analytics" → BigQuery, Dataproc, Dataflow - "Secure" → VPC, IAM, CMEK, Private Google Access

Service Selection Decision Trees¶

Compute Decision:

Stateless application?
├─ YES → Container?
│  ├─ YES → Need Kubernetes?
│  │  ├─ YES → GKE
│  │  └─ NO → Cloud Run
│  └─ NO → Need custom runtime?
│     ├─ YES → App Engine Flexible
│     └─ NO → App Engine Standard
└─ NO → Compute Engine

Database Decision:

Relational needed?
├─ YES → Global transactions?
│  ├─ YES → Cloud Spanner
│  └─ NO → Cloud SQL
└─ NO → Data model?
   ├─ Key-Value/Document → Firestore
   ├─ Wide-Column → Bigtable
   └─ Analytics → BigQuery

Time Management¶

• 120 minutes ÷ 50 questions = 2.4 minutes per question
• First pass: Answer confident questions (60 minutes)
• Second pass: Tackle difficult questions (45 minutes)
• Final pass: Review flagged questions (15 minutes)

Common Traps¶

• ❌ Choosing complex solutions when simple ones suffice
• ❌ Ignoring cost constraints
• ❌ Over-emphasizing technical perfection vs business needs
• ❌ Not considering operational overhead
• ❌ Forgetting about managed service alternatives
• ❌ Mixing up service capabilities and limits

Study Checklist¶

Knowledge Areas: - [ ] Can design multi-tier applications on GCP - [ ] Understand VPC networking, shared VPC, and VPC peering - [ ] Know when to use each compute option (GCE, GKE, App Engine, Cloud Run, Functions) - [ ] Can select appropriate database for use case - [ ] Understand migration strategies and tools - [ ] Know cost optimization techniques - [ ] Can design for high availability and disaster recovery - [ ] Understand IAM, service accounts, and security best practices - [ ] Know monitoring and logging setup - [ ] Familiar with hybrid connectivity options

Preparation: - [ ] Hands-on experience with GCP (build actual projects) - [ ] Review all case studies on exam guide - [ ] Complete practice exams (80%+ score) - [ ] Read official GCP Architecture Framework - [ ] Review common reference architectures - [ ] Practice with GCP Console and gcloud CLI

Pro Tip: The Professional Cloud Architect exam tests your ability to make architecture trade-offs based on business requirements. Always consider: cost, operational overhead, performance, security, compliance, and scalability. The "best" answer balances all these factors based on scenario constraints!

Documentation Count: This fact sheet contains 100+ embedded documentation links to official Google Cloud documentation.

Good luck! This certification demonstrates expert-level cloud architecture skills on Google Cloud Platform.