Introducing the NetApp AFX Systems
Before we begin, . , which include the NetApp AFX 1K model, delivers a unified hardware and software solution that creates a simplified experience specific to the needs of high-performance NAS and S3 customers. One of the key differentiators of the AFX systems is their disaggregated architecture. Unlike traditional systems, they don’t expose the concept of aggregates to the user. The performance and scalability are specifically designed to meet the requirements of AI and ML workloads. AFX provides a simplified user experience enabling it to be administered by both storage specialists and non-specialists.
NetApp Trident and CSI integration
To enable integration with AFX systems, NetApp Trident™ software, NetApp's Container Storage Interface (CSI)–compliant storage orchestrator, now supports provisioning file mode storage using the NFS protocol starting with . The best part is that if you’re an existing Trident customer, you can seamlessly transition to using AFX systems with minimal changes. All you need to do is provide the right credentials in the Trident backend, and everything else is dynamically determined.
Simple provisioning process
Whether you’re a new Trident user or an existing one, the process of provisioning storage with Trident is straightforward. Here are the simple steps.
Note: For existing Trident users, Please upgrade Trident to 25.10. Please find the steps to upgrade Trident here.
- Start by creating a Trident backend with the storage driver ontap-nas. You can do this either by configuring the TridentBackendConfig (TBC) custom resource definition using the Kubernetes-native approach or by using a custom JSON file with tridentctl, a command-line utility for managing Trident. The configuration is similar to any other ontap-nas NFS backend, with the only changes being the user name and password. You can either use cluster management IP with administrator credentials or specify a specific a storage virtual machine (SVM) with its management IP and credentials. For more information, please refer to this document.
# Kubernetes secret required for creating Trident backend from TBC
[root@scs000887790-1 afx]# cat secret.yaml
apiVersion: v1
kind: Secret
metadata:
name: afx-secret
namespace: trident
type: Opaque
stringData:
username: <username>
password: <password>
[root@scs000887790-1 afx]# kubectl create -f secret.yaml
secret/afx-secret created
[root@scs000887790-1 afx]# kubectl get secret afx-secret -n trident
NAME TYPE DATA AGE
afx-secret Opaque 2 20s
# Kubernetes CR TridentBackendConfig (TBC)
[root@scs000887790-1 afx]# cat tbc.yaml
apiVersion: trident.netapp.io/v1
kind: TridentBackendConfig
metadata:
name: afx-backend-tbc
namespace: trident
spec:
version: 1
backendName: afx-backend
storageDriverName: ontap-nas
managementLIF: 1.1.1.1
svm: svm0
credentials:
name: afx-secret
# Or, Trident backend json
[root@scs000887790-1 afx]# cat backend.json
{
"version": 1,
"storageDriverName": "ontap-nas",
"managementLIF": "1.1.1.1",
"backendName": "afx-backend",
"svm": "svm0",
"username": "<username>",
"password": "<password>"
}
# Create Trident Backend via kubectl
[root@scs000887790-1 afx]# kubectl create -f tbc.yaml
tridentbackendconfig.trident.netapp.io/afx-backend-tbc created
[root@scs000887790-1 afx]# kubectl get tbc afx-backend-tbc -n trident
NAME BACKEND NAME BACKEND UUID PHASE STATUS
afx-backend-tbc afx-backend abf2df43-5789-41b0-9b36-d0c1b1660a00 Bound Success
[root@scs000887790-1 afx]# tridentctl get backend afx-backend -n trident
+-------------+----------------+--------------------------------------+--------+------------+---------+
| NAME | STORAGE DRIVER | UUID | STATE | USER-STATE | VOLUMES |
+-------------+----------------+--------------------------------------+--------+------------+---------+
| afx-backend | ontap-nas | abf2df43-5789-41b0-9b36-d0c1b1660a00 | online | normal | 0 |
+-------------+----------------+--------------------------------------+--------+------------+---------+
# Or, create Trident Backend via tridentctl
[root@scs000887790-1 afx]# tridentctl -n trident create backend -f backend.json
+-------------+----------------+--------------------------------------+--------+------------+---------+
| NAME | STORAGE DRIVER | UUID | STATE | USER-STATE | VOLUMES |
+-------------+----------------+--------------------------------------+--------+------------+---------+
| afx-backend | ontap-nas | 66abe907-92c3-4958-99bf-1d5d1856a9bc | online | normal | 0 |
+-------------+----------------+--------------------------------------+--------+------------+---------+
- Step 2: Define a storage class. Create a storage class that corresponds to the type of storage driver you require. This step allows you to define the characteristics of the storage you want to dynamically provision.
[root@scs000887790-1 afx]# cat sc.yaml
apiVersion: storage.k8s.io/v1
kind: StorageClass
metadata:
name: afx-sc
parameters:
backendType: ontap-nas
storagePools: "afx-backend:.*"
provisioner: csi.trident.netapp.io
[root@scs000887790-1 afx]# kubectl create -f sc.yaml
storageclass.storage.k8s.io/afx-sc created
[root@scs000887790-1 afx]# kubectl get sc afx-sc
NAME PROVISIONER RECLAIMPOLICY VOLUMEBINDINGMODE ALLOWVOLUMEEXPANSION AGE
afx-sc csi.trident.netapp.io Delete Immediate false 12s
Now that everything is ready for dynamic provisioning of storage on the AFX system, let's see an example of how to use it.
- Step 3: Create a PVC. Define a PersistentVolumeClaim (PVC) that specifies the amount of storage you need and references the appropriate storage class. This step ensures that your Kubernetes application has access to the required block storage.
[root@scs000887790-1 afx]# cat pvc.yaml
apiVersion: v1
kind: PersistentVolumeClaim
metadata:
name: afx-pvc
spec:
accessModes:
- ReadWriteOnce
resources:
requests:
storage: 1Gi
storageClassName: afx-sc
[root@scs000887790-1 afx]# kubectl create -f pvc.yaml
persistentvolumeclaim/afx-pvc created
Step 4: Confirm PVC binding. After the PVC is created, verify that it’s successfully bound to a persistent volume (PV). This confirmation ensures that the file storage is ready for use by your applications.
[root@scs000887790-1 afx]# kubectl get pvc afx-pvc
NAME STATUS VOLUME CAPACITY ACCESS MODES STORAGECLASS VOLUMEATTRIBUTESCLASS AGE
afx-pvc Bound pvc-0369d050-3229-43fe-8f46-a75bc25acf29 1Gi RWO afx-sc <unset> 24s
Step 5: Use the PVC. Congratulations! You’re now ready to use the PVC in any pod of your choice. Mount the PVC in your pod's specification, and your application will have seamless access to the high-performance file storage provided by the AFX systems using NFS protocol.
[root@scs000887790-1 afx]# cat pod.yaml
apiVersion: v1
kind: Pod
metadata:
name: afx-pod
spec:
containers:
- image: nginx:alpine
name: nginx
volumeMounts:
- mountPath: mnt/pvc
name: local-storage
nodeSelector:
kubernetes.io/arch: amd64
kubernetes.io/os: linux
volumes:
- name: local-storage
persistentVolumeClaim:
claimName: afx-pvc
[root@scs000887790-1 afx]# kubectl create -f pod.yaml
pod/afx-pod created
[root@scs000887790-1 afx]# kubectl get pod afx-pod
NAME READY STATUS RESTARTS AGE
afx-pod 1/1 Running 0 10s
[root@scs000887790-1 afx]# kubectl exec -it afx-pod -- /bin/ash -c "df -H /mnt/pvc"
Filesystem Size Used Available Use% Mounted on
2.2.2.2:/trident_pvc_0369d050_3229_43fe_8f46_a75bc25acf29
1.0G 320.0K 1023.7M 0% /mnt/pvc
Troubleshooting made easy
If you encounter any issues, we've got you covered with some handy commands for troubleshooting. Use kubectl describe on the problematic resource to gather detailed information. For more insights into the system's behavior, you can check the Trident logs by using kubectl or tridentctl.
[root@scs000887790-1 afx]# kubectl get pvc afx-pvc
NAME STATUS VOLUME CAPACITY ACCESS MODES STORAGECLASS VOLUMEATTRIBUTESCLASS AGE
afx-pvc Bound pvc-0369d050-3229-43fe-8f46-a75bc25acf29 1Gi RWO afx-sc <unset> 11m
[root@scs000887790-1 afx]# kubectl describe pvc afx-pvc
Name: afx-pvc
Namespace: default
StorageClass: afx-sc
Status: Bound
Volume: pvc-0369d050-3229-43fe-8f46-a75bc25acf29
Labels: <none>
Annotations: pv.kubernetes.io/bind-completed: yes
pv.kubernetes.io/bound-by-controller: yes
volume.beta.kubernetes.io/storage-provisioner: csi.trident.netapp.io
volume.kubernetes.io/storage-provisioner: csi.trident.netapp.io
Finalizers: [kubernetes.io/pvc-protection]
Capacity: 1Gi
Access Modes: RWO
VolumeMode: Filesystem
Used By: afx-pod
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Normal Provisioning 12m csi.trident.netapp.io_trident-controller-5c487b5b6b-92k5l_130f1ac0-75c2-41be-b0e6-45908828c6b3 External provisioner is provisioning volume for claim "default/afx-pvc"
Normal ExternalProvisioning 12m (x3 over 12m) persistentvolume-controller Waiting for a volume to be created either by the external provisioner 'csi.trident.netapp.io' or manually by the system administrator. If volume creation is delayed, please verify that the provisioner is running and correctly registered.
Normal ProvisioningSuccess 12m csi.trident.netapp.io provisioned a volume
Normal ProvisioningSucceeded 12m csi.trident.netapp.io_trident-controller-5c487b5b6b-92k5l_130f1ac0-75c2-41be-b0e6-45908828c6b3 Successfully provisioned volume pvc-0369d050-3229-43fe-8f46-a75bc25acf29
[root@scs000887790-1 afx]# kubectl get pod -n trident
NAME READY STATUS RESTARTS AGE
trident-controller-5c487b5b6b-92k5l 6/6 Running 0 40m
trident-node-linux-8xfnz 2/2 Running 0 40m
trident-node-linux-pxqn9 2/2 Running 0 40m
trident-node-linux-rq9gc 2/2 Running 0 40m
[root@scs000887790-1 afx]# kubectl logs trident-controller-5c487b5b6b-92k5l -n trident
Defaulted container "trident-main" out of: trident-main, trident-autosupport, csi-provisioner, csi-attacher, csi-resizer, csi-snapshotter
[root@scs000887790-1 afx]# tridentctl -n trident logs
trident-controller log:
time="2025-11-07T06:12:00Z" level=debug msg=Environment PATH="/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin"
time="2025-11-07T06:12:00Z" level=debug msg=Environment HOSTNAME=trident-controller-5c487b5b6b-92k5l
Conclusion
NetApp AFX storage system stands out as a versatile and powerful solution for modern data management needs. Its disaggregated hardware and software architecture is specifically tailored to support high-performance NAS environments, ensuring optimal performance and scalability for AI and ML workloads. This makes AFX an ideal choice for Kubernetes deployments, where the demands of AI and ML applications require robust and efficient storage solutions. Happy configuring, and may your Kubernetes journey be smooth and successful.
