Explanation
Why the waste happens and who it affects.
Once restarted, re-started database instances begin accruing instance-hour charges even if the database is not needed.
Because Aurora does not provide native lifecycle controls to keep clusters stopped indefinitely, this behavior can result in recurring, unintended compute spend - particularly in non-production, seasonal, or infrequently accessed environments where clusters are stopped and forgotten.
Billing model
The pricing dimensions that drive this cost.
Amazon Aurora compute is billed per database instance-hour. When a cluster is stopped, instance-hour charges pause while storage and backup charges continue. Aurora automatically restarts stopped clusters after a 7-day stop period, at which point compute charges resume.
How to detect
3 checks to find it in your estate.
- Review whether stopped Aurora clusters are restarting automatically after the 7-day stop period
- Identify clusters that repeatedly transition from stopped to running without explicit user intent
- Assess whether re-started clusters are serving active workloads or remain unused
How to fix
3 ways to remove the waste.
- Use automation to re-stop clusters in response to automatic restart events, taking care to avoid race conditions by accounting for intermediate database cluster and database instance states
- Delete database instances from idle database clusters, carefully controlling deletion order; add new database instances when the cluster is needed again
- Completely delete idle database clusters to eliminate compute and storage charges, and restore clusters from final snapshots when needed
Documentation
Vendor references for pricing and configuration.