GPU-Nodes-Metrics-Nvidia
使用NVIDIA Data Center GPU Manager (DCGM) dcgm-exporter 通过Prometheus绘制的GPU Nvidia 基础监控信息.
参考:GPU-Nodes-Metrics 12027 需要安装dcgmi,datacenter-gpu-manager_1.7.2_amd64.deb 安装dcgm-exporter,git clone https://github.com/NVIDIA/gpu-monitoring-tools.git
Data source config
Collector config:
Upload an updated version of an exported dashboard.json file from Grafana
Revision | Description | Created | |
---|---|---|---|
Download |