diff --git a/Deployment-to-Kubernetes-and-Minikube.md b/Deployment-to-Kubernetes-and-Minikube.md
index 92cb0da..032ce98 100644
--- a/Deployment-to-Kubernetes-and-Minikube.md
+++ b/Deployment-to-Kubernetes-and-Minikube.md
@@ -257,6 +257,66 @@ section of the operator README.
 > dedicated workers for erasure coding and volume balancing. If you run an admin
 > server, the plugin is the richer option.
 
+## Backup and restore
+
+The Operator can back up a cluster's **filer metadata** as point-in-time
+snapshots and continuously **mirror file data** to cloud object storage or a
+PVC. Destinations, schedules, and mirrors are declared on the `Seaweed`
+resource under `spec.backup`; on-demand backups and restores are separate
+`SeaweedBackup` / `SeaweedRestore` resources. This maps to the underlying `weed`
+primitives: a metadata snapshot is a one-shot `fs.meta.save` (restored with
+`fs.meta.load`), so it is schedulable and retained; data replication is a
+continuous `weed filer.backup`, so it runs as an always-on mirror `Deployment`.
+
+```yaml
+apiVersion: seaweed.seaweedfs.com/v1
+kind: Seaweed
+metadata:
+  name: seaweed1
+spec:
+  image: chrislusf/seaweedfs:latest
+  master: { replicas: 1 }
+  volume: { replicas: 1, requests: { storage: 10Gi } }
+  filer:  { replicas: 1 }
+  backup:
+    storages:
+      s3-main:                              # an S3-compatible destination
+        type: s3
+        s3: { bucket: my-seaweedfs-backups, region: us-east-1 }
+        credentialsSecret: s3-backup-creds  # AWS_ACCESS_KEY_ID / AWS_SECRET_ACCESS_KEY
+    schedule:                               # cron-driven metadata snapshots
+      - name: nightly
+        schedule: "0 2 * * *"
+        storageName: s3-main
+        keep: 7                             # retain the 7 most recent completed snapshots
+    dataMirror:                             # continuous weed filer.backup to the sink
+      - storageName: s3-main
+---
+apiVersion: seaweed.seaweedfs.com/v1
+kind: SeaweedBackup                         # an on-demand metadata snapshot
+metadata: { name: adhoc-1, namespace: default }
+spec:
+  clusterName: seaweed1
+  storageName: s3-main
+---
+apiVersion: seaweed.seaweedfs.com/v1
+kind: SeaweedRestore                        # restore a snapshot back into the cluster
+metadata: { name: restore-1, namespace: default }
+spec:
+  clusterName: seaweed1
+  backupName: adhoc-1
+```
+
+A storage can target S3, Google Cloud Storage, Azure Blob, Backblaze B2, or a
+PersistentVolumeClaim (`type: filesystem`). The leader-elected scheduler creates
+a `SeaweedBackup` each time a cron fires and prunes completed snapshots beyond
+`keep`; restore reconstructs the filer namespace via `fs.meta.load`. When the
+referenced cluster has mTLS enabled, the backup/restore jobs and the mirror
+mount the same `security.toml`/TLS material. For the full field reference,
+storage credential keys, and the metadata-vs-data model, see the
+[Backup & Restore](https://github.com/seaweedfs/seaweedfs-operator/blob/master/BACKUP_SUPPORT.md)
+documentation in the operator repository.
+
 # Recommended: Helm Deployment
 
 The SeaweedFS Helm chart is the most flexible and maintained way to deploy SeaweedFS on Kubernetes.