Added Nvidia_smi compose and readme

This commit is contained in:
bobtiji 2022-01-22 11:54:59 -05:00
parent ccd4833bf4
commit 6e93d47604
2 changed files with 42 additions and 0 deletions

View File

@ -0,0 +1,23 @@
# Prerequisite
NVIDIA container toolkit
sudo apt -y install build-essential nvidia-cuda-toolkit nvidia-headless-495 nvidia-utils-495 libnvidia-encode-495 \
&& distribution=$(. /etc/os-release;echo $ID$VERSION_ID) \
&& curl -s -L https://nvidia.github.io/nvidia-docker/gpgkey | sudo apt-key add - \
&& curl -s -L https://nvidia.github.io/nvidia-docker/$distribution/nvidia-docker.list | sudo tee /etc/apt/sources.list.d/nvidia-docker.list \
&& sudo apt update \
&& sudo apt -y install nvidia-container-toolkit nvidia-container-runtime nvidia-docker2
## Deployment
1. Modify the prometheus configuration template `/etc/prometheus/prometheus.yml` location.
# Job for Nvidia SMI exporter in prometheus config file
- job_name: 'nvidia_smi_exporter'
static_configs:
- targets: ['nvidia_smi_exporter:9835'] # if nvidia_smi_exporter container is not on same docker network , change this line to "- targets: ['whichever ip your host is:9835']"
# Additional Referfences
[Nvidia container toolkit](https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/install-guide.html#install-guide)
[Nvidia GPU exporter Documentation](https://github.com/utkuozdemir/nvidia_gpu_exporter)
[Official Prometheus Documentation](https://prometheus.io/docs/introduction/overview/)
[Some grafana dashboard, not perfect, old, but configurable](https://grafana.com/grafana/dashboards/14574)

View File

@ -0,0 +1,19 @@
---
version: '3'
services:
nvidia_smi_exporter: #To export data from nvidia-smi, needs nvidia-smi and nvidia-container-toolkit installed on host.
image: utkuozdemir/nvidia_gpu_exporter:0.3.0
container_name: nvidia_smi_exporter
runtime: nvidia
environment:
- NVIDIA_VISIBLE_DEVICES=all
- NVIDIA_DRIVER_CAPABILITIES=all
ports:
- "9835:9835"
volumes:
- /usr/bin/nvidia-smi:/usr/bin/nvidia-smi
- /usr/lib/x86_64-linux-gnu/libnvidia-ml.so:/usr/lib/x86_64-linux-gnu/libnvidia-ml.so
- /usr/lib/x86_64-linux-gnu/libnvidia-ml.so.1:/usr/lib/x86_64-linux-gnu/libnvidia-ml.so.1
restart: unless-stopped
#Mount points for volume work on Ubuntu 20.04