Storage Architecture¶

Overview¶

The RH OVE storage architecture provides unified storage management for both container and VM workloads, leveraging Kubernetes-native storage concepts while supporting traditional VM storage requirements.

Storage Components¶

graph TB
    subgraph "Storage Infrastructure"
        A[Storage Classes]
        B[Persistent Volumes]
        C[Container Storage Interface]
        D[Storage Backends]
    end

    subgraph "VM Storage"
        E[DataVolumes]
        F[VM Disks]
        G[CDI - Containerized Data Importer]
        H[VM Templates]
    end

    subgraph "Container Storage"
        I[Persistent Volume Claims]
        J[ConfigMaps]
        K[Secrets]
    end

    A --> B
    B --> C
    C --> D

    E --> G
    F --> E
    H --> E

    I --> B
    J --> K

    B --> E
    B --> I

DataVolume Management¶

DataVolume CRD¶

DataVolumes are the primary mechanism for managing VM storage in RH OVE:

apiVersion: cdi.kubevirt.io/v1beta1
kind: DataVolume
metadata:
  name: web-vm-disk
  namespace: app-web-prod
spec:
  pvc:
    accessModes:
    - ReadWriteOnce
    resources:
      requests:
        storage: 50Gi
    storageClassName: fast-ssd
  source:
    registry:
      url: "docker://registry.redhat.io/rhel8/rhel:latest"

Storage Import Patterns¶

graph LR
    subgraph "Data Sources"
        A[Container Registry]
        B[HTTP Server]
        C[S3 Bucket]
        D[Existing PVC]
    end

    subgraph "CDI Processing"
        E[Import Pod]
        F[Data Processing]
        G[Format Conversion]
    end

    subgraph "Target Storage"
        H[DataVolume]
        I[PVC]
        J[VM Disk]
    end

    A --> E
    B --> E
    C --> E
    D --> E

    E --> F
    F --> G
    G --> H
    H --> I
    I --> J

Storage Classes and Performance Tiers¶

Performance Tiers¶

# High Performance SSD Storage Class
apiVersion: storage.k8s.io/v1
kind: StorageClass
metadata:
  name: high-performance-ssd
provisioner: kubernetes.io/no-provisioner
parameters:
  type: ssd
  iops: "10000"
  throughput: "500Mi"
reclaimPolicy: Delete
volumeBindingMode: WaitForFirstConsumer
---
# Standard HDD Storage Class
apiVersion: storage.k8s.io/v1
kind: StorageClass
metadata:
  name: standard-hdd
provisioner: kubernetes.io/no-provisioner
parameters:
  type: hdd
  iops: "1000"
  throughput: "100Mi"
reclaimPolicy: Retain
volumeBindingMode: WaitForFirstConsumer
---
# Archive Storage Class
apiVersion: storage.k8s.io/v1
kind: StorageClass
metadata:
  name: archive-storage
provisioner: kubernetes.io/no-provisioner
parameters:
  type: archive
  iops: "100"
  throughput: "50Mi"
reclaimPolicy: Retain
volumeBindingMode: WaitForFirstConsumer

VM Disk Configuration¶

VM with Multiple Disks¶

apiVersion: kubevirt.io/v1
kind: VirtualMachine
metadata:
  name: database-vm
  namespace: app-database-prod
spec:
  template:
    spec:
      domain:
        devices:
          disks:
          - name: rootdisk
            disk:
              bus: virtio
          - name: datadisk
            disk:
              bus: virtio
          - name: logdisk
            disk:
              bus: virtio
        resources:
          requests:
            memory: 8Gi
            cpu: 4
      volumes:
      - name: rootdisk
        dataVolume:
          name: db-vm-root
      - name: datadisk
        dataVolume:
          name: db-vm-data
      - name: logdisk
        dataVolume:
          name: db-vm-logs

DataVolume for Different Use Cases¶

# Boot disk from registry
apiVersion: cdi.kubevirt.io/v1beta1
kind: DataVolume
metadata:
  name: db-vm-root
spec:
  pvc:
    accessModes: [ReadWriteOnce]
    resources:
      requests:
        storage: 30Gi
    storageClassName: high-performance-ssd
  source:
    registry:
      url: "docker://registry.access.redhat.com/rhel8/rhel:latest"
---
# Data disk - blank
apiVersion: cdi.kubevirt.io/v1beta1
kind: DataVolume
metadata:
  name: db-vm-data
spec:
  pvc:
    accessModes: [ReadWriteOnce]
    resources:
      requests:
        storage: 500Gi
    storageClassName: standard-hdd
  source:
    blank: {}
---
# Log disk - blank
apiVersion: cdi.kubevirt.io/v1beta1
kind: DataVolume
metadata:
  name: db-vm-logs
spec:
  pvc:
    accessModes: [ReadWriteOnce]
    resources:
      requests:
        storage: 100Gi
    storageClassName: standard-hdd
  source:
    blank: {}

Storage Backup Integration¶

Rubrik Integration for VM Storage¶

Based on our research, Rubrik provides certified integration with RH OVE for VM backup:

graph TB
    subgraph "RH OVE Cluster"
        A[Virtual Machines]
        B[DataVolumes]
        C[Persistent Volumes]
        D[Rubrik Agent]
    end

    subgraph "Rubrik Platform"
        E[Rubrik Cluster]
        F[Backup Policies]
        G[Recovery Points]
        H[Immutable Storage]
    end

    A --> D
    B --> D
    C --> D
    D --> E
    E --> F
    F --> G
    G --> H

Backup Policy Configuration¶

apiVersion: v1
kind: ConfigMap
metadata:
  name: rubrik-backup-policy
  namespace: app-database-prod
data:
  policy.yaml: |
    vm_backup_policy:
      name: "database-vm-backup"
      frequency: "daily"
      retention: "30d"
      snapshot_consistency: "crash-consistent"
      backup_window: "02:00-06:00"
      exclude_disks:
        - "temp-disk"
        - "swap-disk"

Storage Monitoring and Performance¶

Storage Metrics¶

Key storage metrics to monitor:

apiVersion: monitoring.coreos.com/v1
kind: ServiceMonitor
metadata:
  name: storage-metrics
spec:
  selector:
    matchLabels:
      app: cdi-controller
  endpoints:
  - port: metrics
    interval: 30s
    path: /metrics

Performance Monitoring¶

graph LR
    subgraph "Storage Metrics"
        A[IOPS per VM]
        B[Throughput per Volume]
        C[Latency Metrics]
        D[Storage Utilization]
    end

    subgraph "Monitoring Stack"
        E[Prometheus]
        F[Grafana]
        G[Alert Manager]
    end

    A --> E
    B --> E
    C --> E
    D --> E

    E --> F
    E --> G

Storage Operations¶

Volume Expansion¶

# Expand a DataVolume
apiVersion: cdi.kubevirt.io/v1beta1
kind: DataVolume
metadata:
  name: web-vm-disk
  namespace: app-web-prod
spec:
  pvc:
    accessModes: [ReadWriteOnce]
    resources:
      requests:
        storage: 100Gi  # Increased from 50Gi
    storageClassName: fast-ssd
  source:
    pvc:
      name: web-vm-disk
      namespace: app-web-prod

Volume Cloning¶

# Clone a DataVolume for VM template
apiVersion: cdi.kubevirt.io/v1beta1
kind: DataVolume
metadata:
  name: web-vm-template-clone
  namespace: vm-templates
spec:
  pvc:
    accessModes: [ReadWriteOnce]
    resources:
      requests:
        storage: 50Gi
    storageClassName: fast-ssd
  source:
    pvc:
      name: web-vm-golden-image
      namespace: vm-templates

Snapshot Management¶

apiVersion: snapshot.storage.k8s.io/v1
kind: VolumeSnapshot
metadata:
  name: db-vm-snapshot-pre-upgrade
  namespace: app-database-prod
spec:
  volumeSnapshotClassName: csi-snapshotter
  source:
    persistentVolumeClaimName: db-vm-data

Storage Security¶

Encryption at Rest¶

apiVersion: storage.k8s.io/v1
kind: StorageClass
metadata:
  name: encrypted-storage
provisioner: ebs.csi.aws.com
parameters:
  type: gp3
  encrypted: "true"
  kmsKeyId: "arn:aws:kms:region:account:key/key-id"
reclaimPolicy: Delete
volumeBindingMode: WaitForFirstConsumer

Access Control¶

apiVersion: rbac.authorization.k8s.io/v1
kind: Role
metadata:
  namespace: app-database-prod
  name: storage-admin
rules:
- apiGroups: [""]
  resources: ["persistentvolumeclaims"]
  verbs: ["get", "list", "create", "update", "patch", "delete"]
- apiGroups: ["cdi.kubevirt.io"]
  resources: ["datavolumes"]
  verbs: ["get", "list", "create", "update", "patch", "delete"]
- apiGroups: ["snapshot.storage.k8s.io"]
  resources: ["volumesnapshots"]
  verbs: ["get", "list", "create", "delete"]

Best Practices¶

Storage Planning¶

Right-size storage: Match storage performance to workload requirements
Use appropriate storage classes: Different tiers for different use cases
Plan for growth: Consider volume expansion capabilities
Backup strategy: Regular snapshots and external backups

Performance Optimization¶

Use SSD for high-IOPS workloads: Database and application storage
Separate storage by function: OS, data, logs, and temp on different volumes
Monitor storage metrics: Track IOPS, throughput, and latency
Implement storage quotas: Prevent storage exhaustion

Security Considerations¶

Enable encryption at rest: For sensitive data storage
Implement access controls: RBAC for storage resources
Regular security scanning: Check for storage-related vulnerabilities
Audit storage access: Monitor who accesses what storage

This storage architecture ensures reliable, performant, and secure storage management for the hybrid VM and container workloads in the RH OVE environment.