Skip to main content
Fast snapshot restore lets volumes created from a snapshot deliver full performance immediately. It is billed for every hour it is enabled, in every Availability Zone, whether or not any volume is created. nOps finds snapshots with fast snapshot restore enabled around the clock and shows its charge as the saving.

Why it matters

Fast snapshot restore is often turned on for a launch or migration and then forgotten, and its hourly charge per Availability Zone adds up.

How nOps finds it

nOps reads the last 30 complete days of fast snapshot restore billing for each snapshot. A snapshot is flagged when it was billed for at least 20 hours a day on average, and for 20 hours or more on at least 24 days. Snapshots mostly covered by commitments are skipped. The saving is the full on-demand charge over the last 7 days, scaled to a month, and must be at least $5 a month. Billing does not show whether volumes are being restored, so you need to confirm it is not needed.

Before you start

Permission to disable fast snapshot restore (ec2:DisableFastSnapshotRestores).

How to implement

  1. Check whether volumes are created from the snapshot often and need full performance at creation, for example from an Auto Scaling launch template. CloudTrail CreateVolume events for the snapshot show this.
  2. Disable it: EC2 → Snapshots → Actions → Manage fast snapshot restore, or aws ec2 disable-fast-snapshot-restores --source-snapshot-ids ID --availability-zones AZ.
  3. If it was enabled by a script or template, change that too.

Risks and considerations

  • Volumes created from the snapshot afterwards load data lazily, so their first reads are slower.
  • To roll back, enable fast snapshot restore again. It takes some time before it is fully enabled.