GCP Cloud Skill

This skill provides GCP architecture patterns and best practices.

Core Services

Service	Use Case
Compute Engine	Virtual machines
Cloud Functions	Serverless functions
Cloud Run	Containers serverless
Cloud Storage	Object storage
Cloud SQL	Managed databases
BigQuery	Data warehouse
GKE	Kubernetes

Terraform Patterns

# GKE cluster
resource "google_container_cluster" "primary" {
  name     = "my-cluster"
  location = "us-central1"

  remove_default_node_pool = true
  initial_node_count       = 1

  workload_identity_config {
    workload_pool = "${var.project_id}.svc.id.goog"
  }
}

Security Best Practices

• Use Workload Identity (not service account keys)
• Enable VPC Service Controls
• Use Cloud IAM for access management
• Enable Cloud Audit Logs
• Use Customer-Managed Encryption Keys
• Enable Binary Authorization for GKE

BigQuery Patterns

-- Partitioned table for cost optimization
CREATE TABLE mydataset.events
PARTITION BY DATE(event_time)
CLUSTER BY user_id
AS SELECT * FROM staging.events;

Cost Optimization

• Use Committed Use Discounts
• Use Preemptible VMs for batch
• Enable autoscaling
• Use BigQuery slot reservations
• Archive to Coldline/Archive storage