Skip to content

Pricing & Billing

Summary

  • AWS pricing is primarily pay-as-you-go, with options like On-Demand, Reserved Instances/Savings Plans, and Spot to match different workload patterns.
  • Understanding cost drivers (compute, storage, data transfer) and cost management tools is essential to avoid surprises and design cost-efficient architectures.
  • Billing visibility and tagging strategy are as important as technical design for long-term cloud success.

Pricing & billing overview (mindmap)

mermaid
mindmap
  root((Pricing & Billing))
    Pricing Models
      On-Demand
      Reserved Instances
      Savings Plans
      Spot
      Free Tier
    Cost Drivers
      Compute
      Storage
      Data transfer
      Managed services
    Cost Management Tools
      Cost Explorer
      AWS Budgets
      Cost Anomaly Detection
      Cost & Usage Reports
    Allocation
      Tags
      Cost categories
      Accounts/OUs

Best Practices

  • Start with tagging (Environment, Project, Owner, CostCenter) so you can break down costs by workload and team from day one.
  • Use Cost Explorer and AWS Budgets to track trends and set alerts; enable Cost Anomaly Detection for early warning of unusual spend.
  • Mix pricing models: On-Demand for unpredictable workloads, Reserved Instances/Savings Plans for stable baselines, and Spot for fault-tolerant workloads.
  • Understand data transfer pricing (especially cross-Region and internet egress) and design architectures to minimize unnecessary traffic.
  • Review bills monthly and cleanup idle resources (stopped-but-forgotten instances, unattached EBS volumes, unused Elastic IPs, old snapshots).

Exam Notes

  • Be comfortable picking the right pricing model given a scenario (steady vs spiky workloads, tolerance for interruption, long-term commitments).
  • Know which cost tools are used for what: Cost Explorer (analysis), Budgets (alerts), Cost & Usage Reports (detailed data), Trusted Advisor and RI/SP recommendations (optimization).
  • Remember that Free Tier is limited in time and usage; exam scenarios sometimes mention it for small experiments or POCs.

AWS documentation