System Description

Resource Limits per User

These are the default resource limits for each user:

1 CPU core guaranteed (up to 4 possible)
2GB guaranteed system memory (up to 8GB possible)
100GB persistent file storage

Shared GPU Resources

The system includes the following GPU resources that are shared with all users:

4 NVIDIA A100 GPUs, accessible from the PyTorch server

In a notebook or a terminal session a user only has access to part of one GPU.

Notebook Duration Limits

Notebooks are temporary and intended to last a maximum of 24 hours. Notebooks older than 24 hours may be ended without warning.

Notebooks run live in a web browser, if you close your web browser or lose your internet connection for more than 1 hour, your notebook and running processes may end. It is expected that running processes will continue if your connection is interrupted for less than 1 hour.

User Directory

Unlike notebooks, files in your user directory are persistent.

ICRN does not backup data, you are responsible for backing up your files.

Server Options

Python Server

The Python server has Python and Conda installed and runs on a CPU. Python is a general-purpose programming language. Conda is an open source package management system.

R Server

The R server has popular R packages installed and runs on a CPU. R is a programming language and software environment for statistical computing and graphics.

PyTorch Server

The PyTorch server has PyTorch installed and runs on a GPU. NVIDIA provides a brief overview of GPU computing at the beginning of the Using GPUs with Python webinar hosted by NCSA. You can view past recordings of this webinar, and others, in the Past NCSA Training Events Repository in the HPC Moodle.