In clustered Data ONTAP, snapshots are scheduled by creating

Snapshot policies in clustered ONTAP are not created within the vserver (SVM) context, and a vserver-scoped policy that automatically applies to all volumes in that vserver is not how the feature works.

While snapshot policies are applied at the volume level, they are not created within the vserver shell - they must be defined at the cluster admin level (clustershell) to be reusable across SVMs.

Snapshot policies created in the clustershell are available to every SVM and are applied individually at the volume level, meaning each volume must explicitly reference the policy it should use rather than inheriting it automatically.

Because snapshot policies are defined in the clustershell rather than within a specific SVM, each SVM in the cluster can reference and use the same cluster-level policies, eliminating the need for duplicate per-SVM policy definitions.