All Products
Search
Document Center

Container Service for Kubernetes:Update the GPU sharing component

Last Updated:Jul 26, 2024

You can update the GPU sharing component to the latest version in the following case: You have installed the GPU sharing component in your cluster, but the driver version of the GPUs that are used by the nodes is incompatible with that of cGPU in the cluster, or the version of the operating systems that the nodes run is incompatible with the version of the operating system that is used by cGPU in the cluster.

Step 1: Select a method to update the GPU sharing component

You need to select a method to update the ack-ai-installer component based on the method that is used to install ack-ai-installer. You can install the ack-ai-installer component by using one of the following methods:

  • Use the cloud-native AI suite: On the Cloud-native AI Suite page, install the ack-ai-installer component. We recommend that you use this method. For more information, see the Step 1: Install the GPU sharing component section of the "Configure the GPU sharing component" topic.

  • Use application catalog: Install the ack-ai-installer component on the App Catalog tab of the Marketplace page in the Container Service for Kubernetes (ACK) console. This installation method is unavailable. However, you can update the existing ack-ai-installer component that was installed by using this method before.

    Important

    If you uninstalled the ack-ai-installer component that was installed by using this method from a cluster, you must use the cloud-native AI suite to reinstall the component.

How do I check the method that is used to install the GPU sharing component in a cluster?

  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 pane, choose Applications > Cloud-native AI Suite.

  3. Check whether Activate Service is displayed on the Cloud-native AI Suite page.

    If Activate Service is displayed on the Cloud-native AI Suite page, the GPU sharing component is installed on the App Catalog tab of the Marketplace page. Otherwise, the GPU sharing component is installed on the Cloud-native AI Suite page.

Step 2: Update the GPU sharing component

Use the cloud-native AI suite

  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 pane, choose Applications > Cloud-native AI Suite.

  3. In the Components section, find the ack-ai-installer component and click Upgrade in the Actions column.

Use the application catalog

  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 of the cluster, find the ack-ai-installer component and click Update in the Actions column. In the Update Release panel, select the latest chart version and click OK.

    Important

    If you want to modify the configuration of a chart, update the component after you modify the configuration.

    After the component is updated, check the chart version of the ack-ai-installer component on the Helm page.