Administrator Guide

NVDashboard – Nvidia GPU Dashboard
25 RAPIDS Scaling on Dell EMC PowerEdge Servers
E NVDashboard – Nvidia GPU Dashboard
As an alternative monitory tool, NVDashboard is an open-source package for the real-time visualization of
NVIDIA GPU metrics on interactive Jupyter environments. The dashboards use pynvml to access GPU
information attached to the machine and display the plots in Jupyter Lab environment.
Instructions to install NVDashboard on docker
Run the docker image:
$ docker run --gpus all --rm -it --net=host -p 8888:8888 -p 8787:8787 -p 8786:8786
rapidsai/rapidsai:0.10-cuda10.1-runtime-ubuntu18.04
Once inside the docker, add GPU DashBoards with the below commands:
$ pip install jupyterlab-nvdashboard
$ jupyter labextension install jupyterlab-nvdashboard
Start Jupyter notebook, it will run at the designated output IP/Port
$ bash utils/start-jupyter.sh
Once the NVDashboard is installed, the “System Dashboard” will be visible at the left side of the Jupyter-
Lab environment, see below some plot samples