Google Cloud Professional Cloud Architect (PCA) Study Guide 2026

# Google Cloud Professional Cloud Architect (PCA) Study Guide 2026 The Google Cloud Professional Cloud Architect (PCA) is one of the most respected cloud certifications in the industry. It validates your ability to design, develop, and manage robust, secure, scalable, and highly available solutions using Google Cloud. Unlike associate-level exams, the PCA tests architectural judgment—your ability to make the right trade-offs given complex, real-world constraints. The defining feature of the PCA exam is its use of **four published case studies** that candidates can study before exam day. Understanding these scenarios deeply is the fastest path to passing. --- ## Exam Format at a Glance | Detail | Value | |---|---| | Cost | $200 USD | | Number of questions | ~60 | | Duration | 2 hours | | Question types | Multiple choice, multiple select | | Experience recommended | 3+ years in industry, 1+ year in GCP | | Delivery | Online proctored or testing center | | Validity | 2 years | The PCA exam is substantially harder than the ACE. Questions require multi-step reasoning: read a case study requirement, eliminate architecturally incorrect options, and select the answer that best balances all constraints (cost, performance, security, availability). There are often no "obviously wrong" options—all four may be technically valid, but only one meets all the stated requirements. --- ## The Four Case Studies Google publishes four case studies at cloud.google.com/certification/guides/cloud-architect. You should read each one multiple times before exam day. ### 1. Mountkirk Games **Industry**: Online gaming **Core scenario**: Mountkirk Games wants to build a new game backend that can scale globally. They need to support millions of concurrent players with low latency worldwide, analyze player behavior in near real-time, and store game state reliably. **Key requirements**: - Global game backend with low latency for players in multiple regions. - Real-time analytics on player behavior and game events. - Scalable compute for game logic (GPU optional for future game physics). - Cost-effective scaling (scale down when off-peak). **Primary services**: GKE (game backend), Cloud Spanner (globally consistent game state), Pub/Sub + Dataflow + BigQuery (real-time analytics pipeline), Cloud CDN (static assets), Global Load Balancer. ### 2. TerramEarth **Industry**: Heavy equipment manufacturing (IoT) **Core scenario**: TerramEarth manufactures mining and agricultural equipment with embedded sensors. Vehicles transmit telemetry data. TerramEarth wants to predict equipment failures before they occur, analyze vehicle performance, and reduce time-to-repair. **Key requirements**: - Ingest telemetry data from 2 million vehicles (some with intermittent connectivity). - Predictive maintenance using ML. - Secure data transfer from vehicles to cloud. - APIs to share data with dealers and partners. **Primary services**: Pub/Sub (telemetry ingestion), Dataflow (stream processing), BigQuery (analytics), Vertex AI (predictive ML models), Apigee (API management for dealer APIs), Cloud IoT Core (legacy, now Cloud Pub/Sub with MQTT). ### 3. EHR Healthcare **Industry**: Healthcare **Core scenario**: EHR Healthcare is a medical records company looking to migrate from on-premises to Google Cloud. They handle sensitive patient data (HIPAA), need high availability, and want to modernize their legacy application stack. **Key requirements**: - HIPAA compliance (PHI must be protected). - Lift-and-shift existing applications as starting point. - High availability for healthcare applications (no downtime for patient care). - Secure interconnect between on-premises and GCP. - Audit logging for compliance. **Primary services**: Cloud SQL (patient records), Dedicated Interconnect (secure private connectivity), VPC Service Controls (HIPAA boundary), Cloud Audit Logs, Cloud Healthcare API, Identity-Aware Proxy, Cloud KMS (CMEK for PHI encryption). ### 4. Helicopter Racing League **Industry**: Sports media / streaming **Core scenario**: The Helicopter Racing League streams live racing events globally. They want to improve video streaming quality, add real-time race predictions using ML, and expand to new regions. **Key requirements**: - Global video streaming with low latency for viewers worldwide. - Real-time ML predictions during live races (race outcome, leaderboard). - Scalable infrastructure that can handle live event traffic spikes. - Reduce costs by eliminating on-premises broadcast infrastructure. **Primary services**: Cloud CDN (video delivery), Transcoder API (video processing), Pub/Sub + Dataflow (real-time race data), Vertex AI (ML predictions), Global HTTP(S) Load Balancer, Cloud Run (stateless processing services). --- ## Domain Breakdown The PCA exam covers six domains: | Domain | Topic | Approximate Weight | |---|---|---| | 1 | Designing and planning a cloud solution architecture | 24% | | 2 | Managing and provisioning a solution infrastructure | 15% | | 3 | Designing for security and compliance | 18% | | 4 | Analyzing and optimizing technical and business processes | 18% | | 5 | Managing implementation | 11% | | 6 | Ensuring solution and operations reliability | 14% | Domains 1, 3, and 4 together account for 60% of the exam. Mastering architectural design, security, and business process optimization is essential. --- ## Key Concepts to Master ### High Availability and Disaster Recovery | Concept | Definition | |---|---| | HA (High Availability) | Design that eliminates single points of failure; typically multi-zone | | DR (Disaster Recovery) | Plan for recovering from major failures (regional outage) | | RTO (Recovery Time Objective) | Maximum acceptable downtime after a failure | | RPO (Recovery Point Objective) | Maximum acceptable data loss (time since last backup/sync) | DR patterns from cheapest/slowest to most expensive/fastest: | Pattern | RTO | RPO | Approach | |---|---|---|---| | Cold standby | Hours | Hours | Restore from backup when disaster occurs | | Warm standby | Minutes | Minutes | Reduced-capacity standby, scale up on failover | | Hot standby (active-active) | Near-zero | Near-zero | Two active deployments, traffic routed between them | 💡 **Exam Tip:** "Minimize RTO" → hot standby (active-active). "Minimize cost of DR" → cold standby (backup and restore). Match the DR pattern to the business requirement, not the most impressive option. ### SRE Practices Key SRE concepts for the PCA exam: - **Toil**: repetitive, manual, automatable work. SRE practice is to eliminate toil through automation. - **Error budget**: 100% minus the SLO. Used to balance feature velocity vs reliability. - **Burn rate alerting**: alert when the error budget is being consumed faster than expected. - **Postmortem**: blameless analysis after incidents to prevent recurrence. ### VPC Service Controls **VPC Service Controls** creates a security perimeter around Google Cloud API-based services. Resources inside the perimeter can only communicate with other resources inside the same perimeter. This prevents data exfiltration even if IAM credentials are compromised. Key use case: HIPAA and PCI-DSS compliance, where regulated data must not be accessible from outside the organization's GCP environment. ### API Management with Apigee **Apigee** is Google Cloud's full-featured API gateway and management platform. Use it for: - Publishing APIs to external partners (TerramEarth dealer scenario). - Rate limiting, authentication, and analytics for API consumers. - API versioning and lifecycle management. Apigee is the answer whenever a PCA scenario mentions "external API management," "developer portal," or "API monetization." --- ## 8-Week PCA Study Plan | Week | Focus | Activities | |---|---|---| | 1 | Exam overview and case studies | Read all four case studies thoroughly, understand business requirements vs technical requirements | | 2 | Architecture fundamentals | Multi-region design, HA/DR patterns, RTO/RPO, VPC design, Shared VPC | | 3 | Compute and containers | GKE Enterprise, Anthos, Cloud Run, Compute Engine advanced, MIG autoscaling | | 4 | Data and analytics | Cloud Spanner, Bigtable, BigQuery, Pub/Sub, Dataflow, Dataproc | | 5 | Security and compliance | VPC Service Controls, Cloud KMS, Binary Authorization, SCC, IAM deep dive | | 6 | Operations and reliability | SRE, SLO/SLA, Cloud Monitoring, alerting, burn rate, Apigee | | 7 | Case study practice | Apply services to each case study, practice scenario-based questions | | 8 | Full practice exams | CertLand practice exams, review weak areas, exam-day preparation | --- ## What to Expect on Exam Day PCA questions are scenario-heavy and often reference the case studies directly. Expect questions like: - "Mountkirk Games needs to store game state that is consistent across all regions. Which service should they use?" - "EHR Healthcare needs to ensure that HIPAA data cannot leave their GCP perimeter even if credentials are compromised. What should they implement?" - "TerramEarth wants to predict equipment failure 30 days in advance. What is the right approach?" Read each question for all constraints: the business requirement, the technical constraint, the non-functional requirement (cost, security, availability), and which case study applies. The correct answer addresses all constraints—not just the most obvious one. Ready to test your knowledge? [Practice with our Google Cloud Professional Cloud Architect exam](#).