Want to scrape Memory & GPU utilization metrics using NVML & exposes them to Prometheus through a simple HTTP server and/or a push gateway- All in a Docker container???? Then you have come to the right place.
- Install NVIDIA-DOCKER
- A System with GPGPU Card.
Execute the script start.sh
as shown below:
sh start_containers.sh
The above command will start pushgateway, prometheus & Grafana in sequence
python gpu_metrics_exporter.py