Container Service for Kubernetes (ACK) provides the open source cGPU solution that allows you to share one GPU among multiple containers in a Kubernetes cluster. You can enable cGPU for Kubernetes clusters that are deployed on Alibaba Cloud, Amazon Web Services (AWS), Google Compute Engine (GCE), or data centers. cGPU enables GPU sharing to reduce the cost of GPU resources. cGPU also enables the isolation of GPU resources allocated to multiple containers when one GPU is shared. This prevents the issue in which some containers consume excessive resources and other containers run with insufficient resources. In addition, cGPU enables fine-grained sharing of GPUs.
You can refer to the following topics for more details:
For more information about cGPU, see cGPU overview.
For more information about how to enable cGPU Professional Edition, see Configure the GPU sharing component.
For more information about how to use cGPU, see Examples of using cGPU to share GPUs and Work with multiple GPU sharing.
If you migrate workloads from a dedicated Kubernetes cluster installed with cGPU Basic Edition to a professional managed Kubernetes cluster, you must upgrade to cGPU Professional Edition in the professional managed Kubernetes cluster after the migration is completed. For more information, see Update the cGPU component after you migrate from an ACK dedicated cluster to an ACK Pro cluster.