|
|
@@ -0,0 +1,21 @@
|
|
|
+---
|
|
|
+version: '3'
|
|
|
+services:
|
|
|
+
|
|
|
+ nvidia_exporter: #to export data from DCGM host, need DCGM installed of an equal or newer version to the container on the host system https://github.com/NVIDIA/DCGM
|
|
|
+ image: nvcr.io/nvidia/k8s/dcgm-exporter:2.3.2-2.6.2-ubuntu20.04
|
|
|
+ container_name: nvidia_exporter
|
|
|
+ runtime: nvidia
|
|
|
+ cap_add:
|
|
|
+ - SYS_ADMIN
|
|
|
+ environment:
|
|
|
+ - NVIDIA_VISIBLE_DEVICES=all
|
|
|
+ - NVIDIA_DRIVER_CAPABILITIES=all
|
|
|
+ ports:
|
|
|
+ - "9400:9400"
|
|
|
+ restart: unless-stopped
|
|
|
+
|
|
|
+ #NVIDIA Data Center GPU Manager: To export data from DCGM host to prometheus, you need DCGM installed on host as well as nvidia container toolkit
|
|
|
+ #https://github.com/NVIDIA/DCGM
|
|
|
+ #https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/install-guide.html#install-guide
|
|
|
+ #https://github.com/NVIDIA/dcgm-exporter and https://docs.nvidia.com/datacenter/cloud-native/gpu-telemetry/dcgm-exporter.html
|
bobtiji