Software Environment

The containerized deployment of a cluster depends on Kubernetes and MindCluster. You can refer to Table 1 to learn about deployment scenarios and setup tools for information on Kubernetes installation tools. In addition, for details about MindCluster components, see "Introduction" > "Component Description " in MindCluster Cluster Scheduling User Guide.

**Table 1** Dependencies
Dependency	Description	Installed on the Management Node	Installed on the Compute Node
Kubernetes
kubectl	Kubernetes command line tool.	Y	N
kubeadm	Creates and manages Kubernetes clusters.	Y	Y
kubelet	Starts a container on each node in a cluster.	Y	Y
MindCluster
Ascend Device Plugin	Provides device discovery, allocation, and health status reporting functions of Ascend AI Processors based on the Kubernetes device plugin mechanism, to enable Kubernetes to manage Ascend AI Processor resources. This software can be used only after Ascend Docker Runtime is installed.	N	Y
Ascend Operator	Creates a ranktable file and mounts it to a container in ConfigMap mapping mode to implement data communication and task coordination between NPUs on multiple nodes, to optimize the link setup performance for collective communication.	Y	Y
Volcano	Adds features such as affinity scheduling and re-scheduling upon faults based on the open source Volcano scheduling plugin mechanism, to maximize the computing performance of Ascend AI Processors.	Y	Y
Ascend Docker Runtime	Provides Ascend containerization for Docker or containerd to automatically mount required files and device dependencies.	N	Y

Parent topic: Environment Setup