Using Fluid on ARM64 Platform

Prerequisites

Get the latest code of Fluid

The official image for multiple platforms is provided by Fluid community. You can refer to the document to deploy Fluid diretly and jump to the step of Example.

$ mkdir -p $GOPATH/src/github.com/fluid-cloudnative/
$ cd $GOPATH/src/github.com/fluid-cloudnative
$ git clone https://github.com/fluid-cloudnative/fluid.git
$ cd $GOPATH/src/github.com/fluid-cloudnative/fluid

Build AMD64 and ARM64 container images

$ export DOCKER_CLI_EXPERIMENTAL=enabled
$ docker run --rm --privileged docker/binfmt:66f9012c56a8316f9244ffd7622d7c21c1f6f28d
$ docker buildx create --use --name mybuilder
$ export IMG_REPO=<your docker image repo>
$ make docker-buildx-all-push

Get the built image address and image version

Change the image version in helm chart

$ cd $GOPATH/src/github.com/fluid-cloudnative/fluid/charts/fluid/fluid
$ vim values.yaml

Create an ARM64 based Kubernetes cluster and install helm chart

View ARM64 based nodes

kubectl get no -l kubernetes.io/arch=arm64
NAME                       STATUS   ROLES    AGE    VERSION
cn-beijing.192.168.3.183   Ready    <none>   6d3h   v1.22.10
cn-beijing.192.168.3.184   Ready    <none>   6d3h   v1.22.10
cn-beijing.192.168.3.185   Ready    <none>   6d3h   v1.22.10

Installation

$ kubectl create ns fluid-system
$ helm install fluid fluid
NAME: fluid
LAST DEPLOYED: Sat Aug 20 21:43:27 2022
NAMESPACE: default
STATUS: deployed
REVISION: 1
TEST SUITE: None

Example

This example is based on JuiceFSRuntime

1. Creating an ARM64 based Kubernetes cluster

2. Refer to the document to prepare JuiceFS Community Edition

Before using JuiceFS, you need to provide parameters for metadata services (such as Redis) and object storage services (such as MinIO), and create corresponding secrets:

kubectl create secret generic jfs-secret \
    --from-literal=metaurl=redis://redis:6379/0 \
    --from-literal=access-key=minioadmin \
    --from-literal=secret-key=minioadmin

• metaurl: Connection URL for metadata engine (e.g. Redis). Read this document for more information.
• access-key: Access key of object storage.
• secret-key: Secret key of object storage.

Check Dataset to be created

$ cat<<EOF >dataset.yaml
apiVersion: data.fluid.io/v1alpha1
kind: Dataset
metadata:
  name: jfsdemo
spec:
  mounts:
    - name: minio
      mountPoint: "juicefs:///"
      options:
        bucket: "http://minio:9000/minio/test"
        storage: "minio"
      encryptOptions:
        - name: metaurl
          valueFrom:
            secretKeyRef:
              name: jfs-secret
              key: metaurl
        - name: access-key
          valueFrom:
            secretKeyRef:
              name: jfs-secret
              key: access-key
        - name: secret-key
          valueFrom:
            secretKeyRef:
              name: jfs-secret
              key: secret-key
EOF

• mountPoint: Refers to the subdirectory of JuiceFS, which is the directory where users store data in the JuiceFS file system, starts with juicefs://. For example, juicefs:///demo is the /demo subdirectory of the JuiceFS file system.
• bucket：Bucket URL。For example, using S3 as object storage, and the name of the bucket is test.
• storage: Specify the type of storage to be used by the file system, e.g. s3, gs, oss. Read this document for more details.

Attention: Only name and metaurl are required. If the JuiceFS has been formatted, you only need to fill in the name and metaurl.

Since JuiceFS uses local cache, the corresponding Dataset supports only one mount, and JuiceFS does not have UFS, you can specify subdirectory in mountPoint (juicefs:/// represents root directory), and it will be mounted as the root directory into the container.

Create Dataset

$ kubectl create -f dataset.yaml
dataset.data.fluid.io/jfsdemo created

Check Dataset status

$ kubectl get dataset jfsdemo
NAME      UFS TOTAL SIZE   CACHED   CACHE CAPACITY   CACHED PERCENTAGE   PHASE      AGE
jfsdemo                                                                  NotBound   44s

As shown above, the value of the phase in status is NotBound, which means that the Dataset resource is not currently bound to any JuiceFSRuntime resource. Next, we will create JuiceFSRuntime resource.

Check JuiceFSRuntime resource to be create

$ cat<<EOF >runtime.yaml
apiVersion: data.fluid.io/v1alpha1
kind: JuiceFSRuntime
metadata:
  name: jfsdemo
spec:
  replicas: 1
  tieredstore:
    levels:
      - mediumtype: MEM
        path: /dev/shm
        quota: 4Gi
        low: "0.1"
EOF

Create JuiceFSRuntime

$ kubectl create -f runtime.yaml
juicefsruntime.data.fluid.io/jfsdemo created

Check JuiceFSRuntime

$ kubectl get juicefsruntime
NAME      AGE
jfsdemo   34s

Wait a while for the various components of JuiceFSRuntime to start smoothly, and you will see status similar to the following:

$ kubectl get juicefs jfsdemo
NAME      WORKER PHASE   FUSE PHASE   AGE
jfsdemo   Ready          Ready        2m15s

Then, check the Dataset status again and find that it has been bound with JuiceFSRuntime.

$ kubectl get dataset jfsdemo
NAME      UFS TOTAL SIZE   CACHED   CACHE CAPACITY   CACHED PERCENTAGE   PHASE   AGE
jfsdemo   0.00B            0.00B    4.00GiB          0.0%                Bound   3m16s

Check Pod to be create, the Pod uses the Dataset created above to specify the PVC with the same name.

$ cat<<EOF >sample.yaml
apiVersion: v1
kind: Pod
metadata:
  name: demo-app
spec:
  containers:
    - name: demo
      image: nginx
      volumeMounts:
        - mountPath: /data
          name: demo
  volumes:
    - name: demo
      persistentVolumeClaim:
        claimName: jfsdemo
EOF

Create Pod

$ kubectl create -f sample.yaml
pod/demo-app created

Check Pod

$ kubectl get po |grep demo
demo-app                                                       1/1     Running   0          31s
jfsdemo-fuse-fx7np                                             1/1     Running   0          31s
jfsdemo-worker-0                                               1/1     Running   0          10m

You can see that the pod has been created successfully, and the FUSE component of JuiceFS has also started successfully.