Knowledge Base

Below is a list of guides for popular subjects. It is not comprehensive so please search our site if you do not find what you are looking for.

Apptainer is a container platform specifically for HPC.

checkquota is a command for checking your storage quota, requesting quota increases and dealing with large Conda environments.

Connect by SSH provides instruction on how to access the clusters through a command line interface.

Connect by Web provides instruction on how to access the clusters through a web browser.

Custom modules may be used to set your environment.

DDT is a graphical parallel debugger.

Data storage is available on all clusters via various filesystems.

Datasets are available on Della for popular cases like imagenet and alphafold.

Environment modules and custom environment modules make it easy to use software.

Gaussian is software package for quantum chemistry calculations.

Globus can be used to transfer large amounts of data.

GPU Computing covers the hardware at Princeton and related tools

Graphical application instructions for working with visualizations or graphical-user-interface (GUI)s, using virtual desktops or X11 forwarding.

GROMACS is a biomolecular simulation package.

Hugging Face is a machine learning community with a focus on natural language processing.

JAX is a flexible Python library for machine learning research.

Jenkins provides automatic testing typically for continuous integration.

Job Priority of Slurm jobs is explained on this page.

Job Stats or stats.rc provides detailed metrics about running and completed jobs.

Julia is a programming language for scientific and numerical computing.

Jupyter notebooks can be run in various ways on the HPC clusters.

LAMMPS is a molecular dynamics simulation package.

MAP is performance profiler for parallel and GPU codes.

Mathematica is an integrated environment for numerical computing.

MATLAB is a popular environment for scientific computing.

Memory allocation for Slurm script is explained on this page.

mpi4py provides a Python interface to MPI. There are a special set of directions for installing.

NAMD is a molecular dynamics simulation package.

OnDemand Portal details how to use the clusters via a web browser.

Parallel Code give you a basic introduction into the concepts behind parallel programming.

PETSc provides data structures and routines for the scalable solution of scientific applications.

Port Configuration and proxy/default can help in situations where the network defaults are not helpful on our clusters.

Python is the most popular scripting language in the computational sciences.

PyTorch is a popular deep learning framework.

Quantum Espresso is a software suite for electronic structure calculations.

R is popular environment for data analysis, statistics and machine learning.

Recover accidentally deleted files according to this guide.

Research Data Service provide consultations, training, and data curation services to researchers throughout the life cycle of research projects.

Scaling analysis shows how to determine the optimal number of nodes, CPU-cores, etc.

Sharing data with other Research Computing users and external collaborators.

Singularity is a secure alternative to Docker that is specifically for HPC.

Stata is a general-purpose statistical software package.

Slurm is the job scheduler used on the HPC clusters.

Spark is a cluster computing framework for large-scale data processing.

SSH problems are covered on this page.

Supressing Duo is possible using the techniques on this page.

TensorFlow is a popular deep learning framework.

Transfer Files covers how to get files on and off the clusters.

tigress-cifs makes it possibe to mount /tigress and /projects on your local machine.

tigress-web allows one to make files in /tigress available on the internet.

TurboVNC is a graphical desktop sharing system that allows users to connect to remote computers.

VASP is a popular simulation package for ab initio calculations.

If you did not find what your were looking for then please search our site. If that fails then please write to [email protected].