GCDL Cheat Sheet

Key concepts

• ✓Distinguishing CapEx versus OpEx models and how pay-as-you-go pricing shifts IT budgeting
• ✓Mapping Google Cloud products such as BigQuery, GKE and Vertex AI to business outcomes
• ✓Explaining Google Cloud's global regions, zones and edge network as availability and latency advantages
• ✓Recognising how cloud enables faster experimentation, scalability and reduced time-to-market

Watch out for

• ⚠Assuming cloud always costs less than on-premises; total cost of ownership depends on workload, utilisation and egress charges.
• ⚠Confusing a technical feature with a business benefit, choosing an answer that describes configuration rather than the value delivered.
• ⚠Treating digital transformation as purely a technology migration, ignoring cultural, process and organisational change the exam expects.

Key concepts

• ✓Selecting Compute Engine, GKE, Cloud Run, or App Engine for a given workload
• ✓Choosing among Cloud Storage classes, Cloud SQL, Spanner, and BigQuery by use case
• ✓Identifying networking services like VPC, Cloud Load Balancing, and Cloud CDN
• ✓Mapping AI/ML and analytics tools such as Vertex AI and Pub/Sub to needs

Watch out for

• ⚠Confusing Cloud Run with Cloud Functions, or GKE with Compute Engine, when the scenario hinges on containers versus VMs versus event-driven code.
• ⚠Assuming BigQuery is a transactional database; it is an analytics warehouse, so OLTP workloads belong to Cloud SQL or Spanner.
• ⚠Treating Cloud Storage classes as interchangeable, ignoring that Nearline, Coldline, and Archive trade retrieval cost and latency for storage price.

Key concepts

• ✓Selecting Cloud Monitoring alerting policies, uptime checks and dashboards for workload health
• ✓Using Cloud Logging, log sinks and Log Router to route and retain logs
• ✓Applying budgets, alerts and quotas to control cost and prevent resource exhaustion
• ✓Choosing autoscaling (MIG autoscaler, GKE HPA) and SLOs to match demand

Watch out for

• ⚠Confusing Cloud Billing budgets with hard spending caps: budgets only alert, they do not stop resources or block usage automatically.
• ⚠Assuming Cloud Monitoring and Cloud Logging are the same service; logs are stored and queried in Logging, while metrics and alerts live in Monitoring.
• ⚠Treating quotas as fixed limits you cannot change, missing that many quotas are adjustable via console or support requests.

Key concepts

• ✓Applying the shared responsibility model to Google Cloud versus customer-controlled security duties
• ✓Choosing IAM roles, service accounts, and least-privilege policies for a given access scenario
• ✓Selecting encryption options: default Google-managed keys, CMEK via Cloud KMS, or CSEK
• ✓Identifying Security Command Center, Cloud Armor, and Cloud Audit Logs for monitoring and compliance

Watch out for

• ⚠Assuming Google encrypts data at rest by default but believing the customer must still enable encryption manually, when default encryption is automatic.
• ⚠Confusing Cloud IAM, which controls Google Cloud resource access, with Cloud Identity, which manages users and authentication across Google services.
• ⚠Treating the shared responsibility model as fixed, when the customer's security duties shift depending on IaaS, PaaS, or SaaS service type.