From 34c5d447e10c0d110683104a6e4c7ba660cbf5d7 Mon Sep 17 00:00:00 2001 From: bobtiji Date: Sat, 22 Jan 2022 07:49:09 -0500 Subject: [PATCH] first --- .../Nvidia DGCM exporter/docker-compose.yml | 21 +++++++++++++++++++ 1 file changed, 21 insertions(+) create mode 100644 docker-compose/prometheus/exporters/Nvidia DGCM exporter/docker-compose.yml diff --git a/docker-compose/prometheus/exporters/Nvidia DGCM exporter/docker-compose.yml b/docker-compose/prometheus/exporters/Nvidia DGCM exporter/docker-compose.yml new file mode 100644 index 0000000..874136e --- /dev/null +++ b/docker-compose/prometheus/exporters/Nvidia DGCM exporter/docker-compose.yml @@ -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 \ No newline at end of file