Optimize Cloud Spending Across AWS, Azure, and GCP

This skill implements systematic cost optimization strategies to reduce cloud spending across AWS, Azure, and GCP while maintaining performance and reliability, following a four-part framework: visibility (cost allocation tags, cloud cost management tools, budget alerts, cost dashboards), right-sizing (analyzing utilization, downsizing over-provisioned resources, auto-scaling, removing idle resources), pricing models (reserved capacity, spot/preemptible instances, savings plans, committed use discounts), and architecture optimization (managed services, caching, data transfer optimization, lifecycle policies).

For AWS, it covers Reserved Instances (30-72% savings vs on-demand over 1-3 year terms with standard or convertible flexibility), Savings Plans (up to 66-72% savings across EC2/Fargate/Lambda, flexible across instance families/regions/OS), Spot Instances (up to 90% savings for batch/CI/CD/stateless workloads, with a 2-minute interruption notice, best mixed with on-demand for resilience), and S3 lifecycle configuration transitioning objects to cheaper storage classes over time and expiring old data. For Azure, it covers Reserved VM Instances (up to 72% savings, exchangeable), Azure Hybrid Benefit (reusing existing Windows Server/SQL Server licenses for up to 80% savings with RIs), and Azure Advisor recommendations (right-sizing, deleting unused resources, reserved capacity, storage optimization). For GCP, it covers Committed Use Discounts (up to 57% savings on vCPUs/memory), Sustained Use Discounts (automatic, up to 30%, no commitment, for Compute Engine/GKE), and Preemptible VMs (up to 80% savings, 24-hour max runtime, best for batch workloads).

It provides a Terraform-based tagging strategy applying common tags (Environment, Project, CostCenter, Owner, ManagedBy) merged onto resources for cost allocation, and cost monitoring via an AWS Budget resource with a percentage-threshold notification, plus cost anomaly detection tools per provider (AWS Cost Anomaly Detection, Azure Cost Management alerts, GCP Budget alerts).

Its architecture patterns: serverless-first (Lambda/Functions for event-driven workloads, pay only for execution time, no idle costs); right-sized databases per environment (small instances in dev, larger with read replicas in production); multi-tier storage (hot in S3 Standard, warm in Standard-IA after 30 days, cold in Glacier after 90 days, archive in Deep Archive after 365 days); and auto-scaling via a Terraform scaling policy tied to a CloudWatch CPU-utilization alarm.

Its 15-item cost optimization checklist covers tagging, deleting unused resources (EBS, EIPs, snapshots), right-sizing, reserved capacity, auto-scaling, storage class optimization, lifecycle policies, anomaly detection, budget alerts, weekly cost review, spot/preemptible usage, data transfer optimization, caching, managed services, and continuous monitoring. Tools referenced: AWS Cost Explorer/Cost Anomaly Detection/Compute Optimizer, Azure Cost Management/Advisor, GCP Cost Management/Recommender, and multi-cloud tools CloudHealth, Cloudability, and Kubecost.

resource "aws_s3_bucket_lifecycle_configuration" "example" {
  bucket = aws_s3_bucket.example.id

  rule {
    id     = "transition-to-ia"
    status = "Enabled"

    transition {
      days          = 30
      storage_class = "STANDARD_IA"
    }

    transition {
      days          = 90
      storage_class = "GLACIER"
    }

    expiration {
      days = 365
    }
  }
}

Use this skill when reducing cloud spending, right-sizing resources, implementing cost governance, optimizing multi-cloud costs, or meeting budget constraints.

Not for tasks unrelated to cloud cost optimization, or where a different domain or tool is needed. Related skills: terraform-module-library for resource provisioning, multi-cloud-architecture for cloud selection.

Inputs: a cloud environment (AWS, Azure, and/or GCP) and its current spending/utilization data.

Outputs: applied pricing-model optimizations (reserved/spot/committed-use), right-sized resources, storage lifecycle policies, tagging for cost allocation, budget alerts and anomaly detection, and a completed cost optimization checklist.

AWS Cost Explorer, Cost Anomaly Detection, Compute Optimizer; Azure Cost Management, Advisor; GCP Cost Management, Recommender; CloudHealth, Cloudability, Kubecost; Terraform.

Cloud engineers and FinOps teams reducing AWS/Azure/GCP spend through right-sizing, pricing-model selection, tagging, and lifecycle/budget governance.