GCP Cloud Architect study plan

Decompose a fictional business requirement set into functional and non-functional requirements, explicitly separating latency, availability, durability, and compliance constraints

Select GCP compute, storage, and database services for each tier of a three-tier application and justify every choice against a stated constraint

Sketch a VPC topology on paper covering regions, subnets, firewall rules, and private connectivity to on-premises

Identify the decision points where a design choice forces a later provisioning constraint (for example, regional vs. multi-regional storage or zonal vs. regional clusters)

Design an identity and access model that assigns roles at project, folder, and resource level using least privilege, and document at least three cases where a predefined role is insufficient

Diagram a defense-in-depth network security design covering ingress, egress, service-to-service identity, and private access to Google APIs

Classify data in the week-1 design by sensitivity and map each class to encryption, retention, and audit-logging controls

Compare hybrid connectivity options (Cloud VPN, Dedicated Interconnect, Partner Interconnect) on bandwidth, SLA, and cost, and choose one with a written rationale

Provision a VPC with subnets, firewall rules, and a private service access range using Terraform, keeping state remote in a Cloud Storage backend

Write reusable Terraform modules that parameterize environment, region, and machine type so the same code produces dev and prod stacks

Deploy a managed database and a containerized workload from the same configuration and describe how the tool detects and reconciles drift

Compare Terraform, Config Connector, and Deployment Manager on state management, drift handling, and preview capability, and record the tradeoffs

Automate plan review by producing a change summary that a reviewer could approve or reject without reading raw code

Apply the shared responsibility and compliance model to a regulated workload by mapping at least five controls to specific GCP services and audit log sources

Design a logging and monitoring pipeline that captures admin activity, data access, and policy-denied events while staying within a stated retention budget

Build a cost model for the week-1 architecture that breaks spend into per-service monthly estimates and names the top three cost drivers

Propose three concrete cost optimizations (commitments, rightsizing, storage class) and quantify the estimated monthly saving for each

Define at least four budget alerts and quota guardrails that would catch cost anomalies before they exceed a stated threshold

Analyze a legacy on-premises workload and rank migration candidates using a defined framework such as the migration strategies spectrum

Sequence a migration wave plan that accounts for dependencies, downtime tolerance, and rollback options per workload

Design a data migration approach for a relational database covering initial load, change capture, validation, and cutover

Evaluate build-versus-buy and managed-versus-self-managed tradeoffs for two components and document the operational cost of each choice

Define the success metrics and business KPIs that would prove the migration delivered the intended outcome

Build a CI/CD pipeline that promotes a containerized workload from dev to prod with automated tests and a manual approval gate

Configure environment separation using separate projects and folders, and demonstrate that a prod deployment cannot be triggered by a dev credential

Write infrastructure tests or policy constraints that fail the pipeline when a non-compliant resource is proposed

Run a blue-green or canary release on the deployed workload and record the observed traffic split and rollback time

Document the release runbook that another engineer could follow without your involvement

Define SLIs, SLOs, and an error budget for two user journeys and convert them into alerting policies that avoid alert fatigue

Configure dashboards and log-based metrics that surface the top failure modes identified during design review

Write a troubleshooting playbook for three failure scenarios covering diagnosis order, likely root causes, and mitigation steps

Design a capacity and autoscaling plan that states the scaling signal, minimum and maximum bounds, and the expected behavior under a traffic spike

Review the full architecture for single points of failure and document the resilience gap and remediation for each finding

Re-audit the full solution against every exam domain and list the weakest three areas with a remediation action for each

Complete two timed practice sets and categorize every wrong answer by domain and by failure type (knowledge gap, misread, or design judgment)

Re-derive the core design decisions from memory without notes to confirm retention of service tradeoffs

Rewrite the week-1 design document incorporating everything learned, and diff it against the original to confirm growth in architectural judgment