NVIDIA GPUs are used to accelerate scientific computation and graphics rendering. Container Service for Kubernetes (ACK) supports centralized scheduling and management of various NVIDIA GPU models. This improves the utilization of GPU resources in ACK clusters. This topic describes how to create an ACK cluster with GPU-accelerated nodes and how to view GPUs that are attached to GPU-accelerated nodes.
Prerequisites
An ACK Pro cluster or an ACK dedicated cluster is created.
Create a GPU-accelerated node pool
Select GPU-accelerated instance types when you create a node pool. For more information, see Create a node pool. For more information about GPU-accelerated instance types, see GPU-accelerated ECS instance types supported by ACK.
If no GPU-accelerated instance type is available, change the specified vSwitches and try again.
View GPUs that attached to GPU-accelerated nodes
After you create an ACK cluster with GPU-accelerated nodes, you can view GPUs that are attached to GPU-accelerated nodes in the cluster.
Log on to the ACK console. In the left-side navigation pane, click Clusters.
On the Clusters page, click the name of the cluster that you want to manage and choose in the left-side navigation pane.
On the Nodes page, select a worker node and choose in the Actions column. On the node details page, you can view the GPUs that are attached to the node.