Welcome to the BGU HPC Cluster

BGU HPC Cluster Documentation

This guide helps you effectively utilize the Ben-Gurion University of the Negev’s High-Performance Computing (HPC) cluster. Whether you’re a seasoned researcher or new to HPC, you’ll find valuable resources here.

This guide shows researchers how to harness the immense computational power of our cluster for advanced research and development. Whether you’re analyzing large datasets, simulating complex systems, or performing machine learning experiments, this cluster helps you push the boundaries of your work. With a scalable infrastructure, support for various scientific tools, and an efficient scheduling platform, the BGU HPC cluster offers a reliable environment tailored to meet diverse research needs.

What is HPC?

High-Performance Computing (HPC) is the use of powerful, multi-node computing systems capable of processing vast amounts of data and solving computational problems at high speeds. Unlike traditional computing, where tasks are handled by a single machine, HPC distributes workloads across many interconnected computers (nodes), allowing researchers to run highly parallelized tasks efficiently. This capability is crucial in fields such as bioinformatics, physics, chemistry, engineering, and data science, where solving large-scale problems quickly can mean the difference between weeks of waiting and actionable results in hours.

Run:AI - Our Scheduler and Platform

We use Run:AI to manage and optimize workloads on the BGU HPC cluster. Run:AI is a modern platform for orchestrating and scheduling AI and compute workloads. It simplifies job scheduling by dynamically allocating resources based on the real-time needs of users. This frees researchers from manually managing resources or waiting in long queues. By abstracting away much of the complexity of traditional HPC scheduling, Run:AI ensures that computational resources are used efficiently, enabling faster job completion and greater flexibility. Whether you’re running a single task or managing multiple concurrent workloads, Run:AI makes the process seamless and straightforward.

Supported Languages, Applications, and IDEs

Our HPC environment supports a wide range of programming languages, applications, and Integrated Development Environments (IDEs) commonly used by researchers. Supported tools include:

  • GROMACS and NAMD, popular molecular dynamics simulation packages widely used in computational chemistry and biophysics.

  • Stata X11, Stata Broswer, R, and RStudio, which support statistical computing and data analysis for social sciences, epidemiology, and other data-heavy disciplines.

  • Python, Jupyter, and Julia, enabling users to perform data analysis, machine learning, and scientific computing interactively.

  • Matlab, a programming and numeric computing platform for engineering and scientific applications like data analysis, signal and image processing, control systems, wireless communications, and robotics.

  • VSCode, PyCharm, and Jupyter Notebook, widely used IDEs that provide robust support for Python development, making it easy to write, debug, and manage code on the cluster.

  • SUMO simulation tool used to model and analyze urban mobility, including road vehicles, public transport, pedestrians, and cargo logistics.

  • Wolfram Mathematica, a software system and programming environment used for technical computing across scientific and technical disciplines, enabling users to perform symbolic computation, data analysis, machine learning, visualization, and numerical simulations.

  • Gurobi, used to solve complex real-world problems using mathematical optimization, providing data-driven, optimal solutions for decision-making across industries like supply chain, finance, manufacturing, and machine learning.

  • OneAPI, a unified software programming model designed to enable developers to create high-performance applications that can run efficiently across a variety of hardware architectures. This includes CPUs, GPUs, FPGAs, and other specialized accelerators.

  • ImageJ, a wide range of tasks in scientific image analysis, including editing, processing, analyzing, quantifying, and visualizing image data.

  • Universal Code Image, supports any programming language that works with Conda environments.

This comprehensive range of tools ensures that no matter your discipline or preferred workflow, you can find the right resources to accelerate your research. We are committed to maintaining an up-to-date environment and will support additional tools as research needs evolve.

Get Started!

This guide walks you through everything you need to know—from getting started with the cluster, submitting jobs, and managing your workloads, to best practices for using specific tools like Jupyter Notebooks or IDEs like VSCode and PyCharm. We have designed this guide to be modular, allowing you to jump directly to sections relevant to your research or workflow. Whether you’re a new user or an experienced HPC researcher, we hope this guide makes your experience on the BGU HPC cluster productive and enjoyable. Happy computing!