Glossary

Cloud automation

Cloud automation uses code, templates and policies to provision, configure and manage cloud resources without manual steps. Networks, storage, compute and permissions are described once and built identically every time, and each change is recorded, reviewed and repeatable in any environment.

The cycle behind every change

  1. Describe the target state in code: networks, storage allocations, accounts and permissions.
  2. Commit the code to version control, so each change has an author, a date and a review.
  3. Let the tool compare the code with what is running and produce a plan listing the differences.
  4. Apply the plan, making only the changes needed to close the gap.
  5. Compare again on a schedule to catch anything changed outside the code.

Because the same code produces the same result, a second region, a test account or a rebuilt environment matches the original rather than approximating it. Two people following one runbook by hand tend to produce two environments that differ by a port or a retention setting, and the difference surfaces months later as an outage in only one of them.

Drift starts with a console fix

A change made outside the code is drift. It usually happens during an incident, when someone corrects a setting by hand and moves on. The next deployment then either reverses the fix or fails. Drift detection reports resources that no longer match, and each one is settled by reverting it or by writing the change back into the code. Automating a process that is already inconsistent only repeats the inconsistency faster, so the target state is agreed before it is coded.

Controls that attach to code

Once resources are defined in code, other rules ride along: mandatory cost tags for cloud cost management, checks that block a publicly readable bucket, and self-service templates that replace tickets. Nothing restricts this to public clouds. Any system with an API can be driven the same way, which is how a private cloud offers self-service and how software-defined storage is provisioned. Routine operations on storage systems themselves are covered under storage automation.