This wiki site will be offline Weds, July 6th, 2022, from 5:30-8:30 PM CDT in order to upgrade Confluence
Child pages
Skip to end of metadata
Go to start of metadata

In publications and presentations that use results obtained on this system, please include the following acknowledgement: “This work utilizes resources supported by the National Science Foundation’s Major Research Instrumentation program, grant #1725729, as well as the University of Illinois at Urbana-Champaign”.

System Description

Host name:



  • CentOS 7
  • CUDA 11
  • NCCL
  • cuDNN
  • TensorRT

User Guide

To request an account: email

Use /home/<username> for basic stuff only, do not put any code/data here as the /home partition is very small

Use /raid partition for all your data/code. Create a subfolder in this partition for your project and keep your stuff there.

To access the system, submit jobs, etc.: Access hal-dgx and overdrive with hal-login3 Node

To compile CUDA applications, first enable devtoolset 9

scl enable devtoolset-9 bash

To work with Python, create and enable virtual environment first

python3 -m venv python3-virtualenv
source python3-virtualenv/bin/activate

You can use pip to install python packages within this environment.

To start a Jupyter notebook on hal-dgx

  • on hal-dgx:
source python3-virtualenv/bin/activate
pip install jupyter # this needs to be done only once
jupyter notebook  # this will start the jupyter server
  • on your own computer:
# This opens a connection to the Jupyter server, and 
# forwards any connection to port 8888 on the local machine to port 8888 on
ssh -L 8888:localhost:8888
  • Finally, on your own computer, open web browser and point it to the address you see after running 'jupyter notebook' on hal-dgx, something like http://localhost:8888/?token=...

Main -> Systems -> DGX A100

Contact us

Request access to ISL resources: Application

Contact ISL staff: Email Address

Visit: NCSA, room 3050E

  • No labels