Use a statically provisioned volume of CPFS General-purpose Edition - Container Service for Kubernetes

Cloud Parallel File System (CPFS) provides high throughput and high input/output operations per second (IOPS). It is suitable for high-performance computing (HPC) scenarios, such as AI training, autonomous driving, gene computing, and video rendering. This topic describes how to create a statically provisioned volume of CPFS General-purpose Edition and mount it to a workload.

Prerequisites

Ensure that the csi-plugin and csi-provisioner components are version v1.22.11-abbb810e-aliyun or later. To upgrade the components, see Upgrade CSI components.

Limits

CPFS General-purpose Edition is available only in specific regions. For more information, see Available regions.
Only the NFS protocol is supported for mounting. The POSIX protocol is not supported.
CPFS supports only nodes that use the x86 architecture.
Volumes can be mounted only to clusters that are in the same VPC.
You cannot mount volumes to nodes that run the ContainerOS operating system.

Configure storage components

Choose a configuration method based on the version of the csi-plugin component.

csi-plugin version 1.33 or later
Install the cnfs-nas-daemon component and configure the csi-plugin to enable the AlinasMountProxy=true FeatureGate. This allows the CSI component to call cnfs-nas-daemon for mounting. For more information, see Manage the cnfs-nas-daemon component.

csi-plugin version earlier than 1.33

Configure a ConfigMap to enable the NFS protocol for mounting CPFS General-purpose Edition.

cat << EOF | kubectl apply -f -
apiVersion: v1
kind: ConfigMap
metadata:
  name: csi-plugin
  namespace: kube-system
data:
  cpfs-nas-enable: "true"   # Use the NFS protocol to mount CPFS General-purpose Edition.
EOF

Restart the csi-plugin to install related dependencies.
This operation does not affect your services.
```
kubectl -n kube-system rollout restart daemonset csi-plugin
```
Expected output:
```
daemonset.apps/csi-plugin restarted
```

Step 1: Create a CPFS file system and a protocol service

CPFS supports only static mounting. You must create a CPFS file system and a corresponding protocol service, and then obtain the mount address to use for mounting.

Create a CPFS General-purpose Edition file system.
- Create the CPFS file system in the same region as your cluster.
- If you use an existing CPFS file system, go to the NAS console. On the File System List page, click the target CPFS file system. On the Basic Information page, verify that the CPFS version is 2.3.0 or later. If the version is earlier than 2.3.0, create a new CPFS file system.
Create a protocol service.
- Use the VPC and vSwitch of the cluster to create the protocol service and generate a mount target.
- If you use an existing protocol service, make sure that it uses the same VPC as the cluster. CPFS does not support cross-VPC mounting. If the protocol service is in a different VPC, you must create a new protocol service.
Obtain the mount address of the protocol service.
On the Protocol Service page, click Export Directory. In the Mount Address column, copy the mount address and save it. The address consists of the mount target domain name and the mount directory. For example, if the exported directory is / , the mount address is displayed as cpfs-****.<Region ID>.cpfs.aliyuncs.com:/share.

Step 2: Create a PV and a PVC

Create a persistent volume (PV) and a persistent volume claim (PVC) based on the existing CPFS file system.

Save the following content as cpfs-pv-pvc.yaml.

apiVersion: v1
kind: PersistentVolume
metadata:
  name: cpfs-pv
  labels:
    alicloud-pvname: cpfs-pv
spec:
  accessModes:
  - ReadWriteMany
  capacity:
    storage: 20Gi
  csi:
    driver: nasplugin.csi.alibabacloud.com
    volumeAttributes:
      mountProtocol: cpfs-nfs                                         # Mounts the file system over the NFS protocol.
      path: "/share"                                                  # The directory in the mount address.
      volumeAs: subpath                                               # A subdirectory.
      server: "cpfs-******-******.cn-shanghai.cpfs.aliyuncs.com"      # The domain name in the mount address.
    volumeHandle: cpfs-pv                                             # Must be the same as the PV name.
  mountOptions:
  - rsize=1048576,wsize=1048576,hard,timeo=600,retrans=2,noresvport
  - vers=3
---
apiVersion: v1
kind: PersistentVolumeClaim
metadata:
  name: cpfs-pvc
spec:
  accessModes:
  - ReadWriteMany
  resources:
    requests:
      storage: 20Gi
  selector:
    matchLabels:
      alicloud-pvname: cpfs-pv

PV parameters

Parameter	Description
`labels`	Set a label. This allows a PVC to use a `selector` to match and attach to the PV.
`accessModes`	The access mode of the PV.
`capacity.storage`	The storage capacity of the volume.
`csi.driver`	The driver type. Set it to `nasplugin.csi.alibabacloud.com`.
`csi.volumeAttributes`	The properties of the CPFS volume. `mountProtocol`: Set to `cpfs-nfs`. This indicates that CPFS is mounted over the NFS protocol. `path`: Set to the path of the exported directory of the CPFS General-purpose Edition protocol service, such as `/share`. You can also set it to a subdirectory, such as `/share/dir`. `volumeAs`: Set to `subpath`. This indicates that a subdirectory-type PV is created. `server`: Set to the domain name in the mount address of the exported directory of the CPFS General-purpose Edition protocol service.
`csi.volumeHandle`	Must be the same as the PV name.

PVC parameters

Parameter	Description
`accessModes`	The access mode that the PVC requests from the PV.
`selector`	Uses the label on the PV to match and attach.
`resources.requests.storage`	The storage capacity allocated to the pod. It cannot be larger than the PV capacity.

Create the PV and PVC.
```
kubectl apply -f cpfs-pv-pvc.yaml
```

Confirm that the PVC is bound to the PV.

kubectl get pvc cpfs-pvc

The following example shows a sample output:

NAME       STATUS   VOLUME    CAPACITY   ACCESS MODES   STORAGECLASS   VOLUMEATTRIBUTESCLASS   AGE
cpfs-pvc   Bound    cpfs-pv   20Gi       RWO                           <unset>                 18m

Step 3: Create an application and mount the CPFS volume

Create a StatefulSet and mount the statically provisioned volume of CPFS General-purpose Edition.

Save the following content as cpfs-test.yaml.

apiVersion: apps/v1
kind: StatefulSet
metadata:
  name: cpfs-sts
spec:
  selector:
    matchLabels:
      app: nginx
  serviceName: "nginx"
  replicas: 1
  template:
    metadata:
      labels:
        app: nginx
    spec:
      containers:
      - name: nginx
        image: anolis-registry.cn-zhangjiakou.cr.aliyuncs.com/openanolis/nginx:1.14.1-8.6
        volumeMounts:
        - name: cpfs-pvc
          mountPath: /data
      volumes:
      - name: cpfs-pvc
        persistentVolumeClaim:
          claimName: cpfs-pvc

Create the StatefulSet.
```
kubectl apply -f cpfs-test.yaml
```

Confirm that the statically provisioned volume of CPFS General-purpose Edition is mounted.

kubectl exec cpfs-sts-0 -- mount | grep /data

The following output indicates that the statically provisioned volume of CPFS General-purpose Edition is successfully mounted.

cpfs-******-******.cn-shanghai.cpfs.aliyuncs.com:/share on /data type nfs (rw,relatime,vers=3,rsize=1048576,wsize=1048576,namlen=255,hard,nolock,noresvport,proto=tcp,port=30000,timeo=600,retrans=2,sec=sys,mountaddr=127.0.1.255,mountvers=3,mountport=30000,mountproto=tcp,local_lock=all,addr=127.0.1.255)