Component Installation Positions

Table 1 shows the installation positions of the components.

**Table 1** Component installation positions
Node	Component	Function
Master node	HCCL-Controller	A plug-in developed based on the Kubernetes informer mechanism. It automatically generates the HCCL configuration file (ranktable file, also called the hccl.json file) for NPU training jobs.
	Volcano	Enhances the affinity scheduling function of AI Processors based on the cluster scheduling function of the open source Volcano.
	Resilience-Controller	Provides resilience control for the minimum training system. When the hardware used by a training job is faulty, the hardware is removed to continue the training.
Worker node	Ascend Device Plugin	Provides the common device plugin mechanism and standard device APIs for Kubernetes to use devices.
	NPU-Exporter	Monitors Ascend AI Processors and container-level allocation status based on Prometheus.
	NodeD	Provides node monitoring functions, such as node heartbeat reporting.
	Ascend Docker Runtime	Provides container-based Ascend NPUs (Ascend AI Processors) for all AI jobs so that AI jobs can smoothly run on Ascend devices as Docker containers.
Training container	Elastic-Agent	Provides functions such as the dying gasp (the last CKPT file) for resumable training and restoration policy in data parallel and hybrid parallel scenarios. To use the dying gasp function, you need to install this component in the training container.

Parent topic: References