Cloud cost optimization exercise
A structured pass through cloud spend — tagging, rightsizing and idle-resource cleanup — with the reasoning behind each decision.
The problem
Cloud cost optimization is often treated as a one-time cleanup rather than an engineering discipline. This lab works through a realistic optimization pass end to end, from visibility to enforced tagging to rightsizing decisions.
Architecture
- A tagging standard enforced through Terraform and policy checks, not manual convention
- Cost allocation dashboards broken down by team, service and environment
- A rightsizing pass based on actual CPU, memory and network utilization data
- Automated detection of idle and orphaned resources (unattached volumes, idle load balancers)
Technologies
Design decisions
Enforce tags at the infrastructure-as-code level, not through manual tagging
Manual tagging discipline degrades within weeks. Enforcement at the Terraform level doesn't.
Rightsize based on two weeks of real utilization data, not defaults
Default instance sizes are guesses. Real usage data turns rightsizing into a measurable decision.
Lessons learned
- Visibility has to come before optimization — you can't rightsize what you can't see
- The biggest savings are usually in forgotten resources, not in aggressive rightsizing of active ones
Working through something similar?
This lab reflects our own reference work, not a client engagement — but the approach transfers directly.
Start a technical conversation