[Component upgrade] shared GPU scheduling component upgrade notice

Affected shared GPU scheduling component versions

Affected existing GPU nodes

Existing GPU nodes are affected if their cGPU version is less than 1.5.13. To check a node’s cGPU version, perform the following steps:

1. Log on to the node.

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

   2. On the Clusters page, click the name of your cluster. In the left navigation pane, click Nodes > Node Pools.

   3. On the Node Pools page, click the node pool name. Then, access node management and log on to the target GPU node.

      ContainerOS does not support direct user logon by default. This prevents untraceable operations and eliminates SSH access to reduce security risks. If you must log on to a ContainerOS instance for O&M tasks, see O&M ContainerOS nodes.

2. Run the following command to obtain the cGPU version:

   cat /proc/cgpu_km/version

   Example output:

   1.5.12

Upgrade the shared GPU scheduling component