Container Service for Kubernetes:Deploy the cloud-native AI suite in an ACK Edge cluster

Prerequisites

• An ACK Edge cluster that runs Kubernetes 1.18 or later is created. For more information, see Create an ACK Edge cluster in the console.

• The csi-plugin and csi-provisioner components are installed in the ACK Edge cluster. For more information, see Manage components.

Step 1: Deploy the NGINX Ingress controller

1. Log on to the ACK console. In the left-side navigation pane, click Clusters.

2. On the Clusters page, find the cluster that you want to manage and click its name. In the left-side navigation pane, choose Applications > Helm.

3. On the Helm page, click Deploy. In the Basic Information step, configure the parameters based on the following table.

   Parameter	Example
   Application Name	ack-ingress-nginx-{Node pool name}
   Namespace	kube-system
   Source	Default value: Marketplace.
   Chart	Use Scenarios: Select All. Supported Architecture: Select amd64. Search box: Search for ack-ingress-nginx-v1.

4. Click Next. On the Parameters page, configure the parameters. The following table describes the parameters. Click OK to complete the installation.

   Note

   Log on to the ACK console. In the left-side navigation pane, choose Nodes > Node Pools to view the node pool list and find the ID of the node pool.

   • Add the following label to the service. nodeSelector parameter: alibabacloud.com/nodepool-id {Node pool ID}.

     Note

     You need to specify only alibabacloud.com/nodepool-id: {Node pool ID} in the service. nodeSelector parameter.

     

   • Set the name and controllerValue fields in the ingressClassResource parameter.

     Note

     If you want to deploy ack-ingress-nginx-v1 in an ACK Edge cluster multiple times, you must specify unique values for the name and controllerValue fields each time you deploy ack-ingress-nginx-v1. Set the fields based on the following description:

     • Set name in the ack-nginx-{Node pool name} format. Example: ack-nginx-edge-hangzhou.

     • Set controllerValue in the "k8s.io/ack-ingress-nginx-{Node pool name}" format. Example: "k8s.io/ack-ingress-nginx-edge-hangzhou".

     

   • On the Parameters page, select an internal-facing or Internet-facing Server Load Balancer (SLB) instance by specifying the value of the enabled parameter below the external and internal parameters. If you select an Internet-facing SLB instance, resolve the domain name to a public IP address. If you select an internal-facing SLB instance, use an accessible and resolvable private IP address.

     

Step 2: Deploy the cloud-native AI suite

For more information about how to deploy the cloud-native AI suite, see Deploy the cloud-native AI suite. Take note of the following information during the installation.

• We recommend that you deploy the cloud-native AI suite on on-cloud nodes to use the cloud computing capabilities. If you want to schedule the cloud-native AI suite to a specific node pool, configure the Selector or Affinity parameter of the corresponding pod.

• If you want to use a self-managed data storage service in your cluster, you must install the ack-mysql component when you deploy the cloud-native AI suite. The component must be deployed on an on-cloud node.

  After the installation is complete, two Ingresses are displayed on the Ingresses page. You must add the ingressClassName parameter to ensure that the Ingress Class is the same as the Ingress Class of the deployed NGINX Ingress controller.

  1. In the left-side navigation pane, choose Network > Ingresses. On the Ingresses page, find the Ingress that you want to manage and click Edit YAML in the Actions column.

  2. Add the ingressClassName parameter and make sure that the value of this parameter is the same as the value of the ingressClassResource.name parameter in Step 1.

• If you want to access the console through a public domain name during the deployment, resolve the domain name to the IP address of the NGINX Ingress controller that you installed. In the left-side navigation pane, choose Network > Services to view the IP address.

After the deployment is complete, you can access the console by using AI Dashboard. For more information, see Access AI Dashboard.