Why it matters
Full snapshots are the default in older scripts and templates, and each one is a full-price copy of the disk.How nOps finds it
nOps reads your pay-as-you-go managed disk snapshot billing. A snapshot is flagged when it was billed on the Premium SSD or Standard SSD snapshot price on at least 7 of the last 14 complete days. The saving is its cost over the last 7 days, scaled to a month, times the difference between the Standard HDD snapshot list price and its current list price in the same region and redundancy. Incremental snapshots usually cost even less, so this is a lower bound. The Standard HDD price must be at least 10% lower, and the saving at least $5 a month. Snapshots old enough to delete are left to that recommendation instead.Before you start
Permission to create and delete snapshots, and access to the source disk.How to implement
- Check the snapshot is still needed and that its source disk still exists.
- Take an incremental snapshot of the source disk:
az snapshot create --resource-group MY_RG --name MY_SNAPSHOT --source DISK_ID --incremental true, or in the portal choose Create snapshot and set Snapshot type to Incremental. - Delete the full snapshot (
az snapshot delete --resource-group MY_RG --name OLD_SNAPSHOT). - Update the scripts, pipelines or templates that create snapshots so they use incremental snapshots.
Risks and considerations
- An incremental snapshot needs its source disk. If the disk has been deleted, keep the full snapshot or delete it if not needed.
- Deleting the full snapshot is permanent, so take the incremental snapshot first.
FAQs
Are incremental snapshots slower to restore?
Are incremental snapshots slower to restore?
You restore a disk from an incremental snapshot the same way as from a full one. Check the restore time for your disk size before relying on it.