Skip to content

The HBS Grid makes massive computing power accessible to the whole HBS research community. Our environments offer a familiar desktop interface and configured software applications including Julia, Jupyterlab, Matlab, Python, R / Rstudio, Stata, VSCode, and hundreds of other popular programs

The HBS Grid consists of many computers connected by a fast network and shared network storage. These powerful machines have up to 1.4Tb of memory and up to 32 cores each, and together they provide the backbone of HBS's research computing capabilities. User-friendly tools and environments are built on top of this physical and network infrastructure, making the power of our computing cluster accessible to all HBS researchers. Using the HBS Grid you can:

  • Interactively analyze data too large for local memory.
  • Speed up your analysis using hundreds of CPUs across multiple machines.
  • Access your persistent remote desktop from anywhere with an internet connection.
  • Store, backup, and access large research data.
  • Share data and collaborate with HBS affiliates and guests.

Quick start

HBS faculty, staff, doctoral students and their guests are eligible for HBSGrid accounts. You can setup your account and connect via either GUI or terminal following the instructions below.

Getting Started

  1. If you do not yet have an HBS Grid account request one here.
  2. Connect to the HBS network, either directly if you are on-campus or connect via VPN otherwise.
  1. If the NoMachine application is not yet installed download and install it.
  2. Start the NoMachine application.
  3. If this is your first time connecting, click the add button and enter these connection details:
    • Name: "HBS Grid", or anything you like
    • Host: hbsgrid-nx.hbs.edu
    • Port: 4000
    • Protocol: NX
  4. Click the connect button and enter your Username and Password.

The video below demonstrates these steps visually.

NoMachine maintains a detailed connection guide that you can refer to if needed. To troubleshoot connection difficulties, see our troubleshooting checklist.

A selection of our most popular applications are available in the favorites list pinned to the task-bar. Additional application launchers can be found in the Applications menu or by searching in Activities. You can add applications to your favorites list by right-clicking and selecting Add to Favorites.

This environment was designed to be intuitive and user-friendly, and you are encouraged to start exploring the available software and tools.

  1. In your favorite terminal program, run 

    ssh USERNAME@hbsgrid.hbs.edu
    and enter your password when prompted.

    • If you are on a PC and choose to use SecureCRT, click "Quick connect..." and enter the following:
      • Protocol: SSH2
      • Hostname: hbsgrid.hbs.edu
      • Port: 22
      • Firewall: none
      • Username: your HBSGrid username

  2. The software environment is not activated by default when connecting via ssh. Run 

    ml rcs
    to enable the software environment.

Next Steps

To work on the cluster, you will probably want to learn about HBSGrid storage and transfer your data. You may also refer to our documentation on launching jobs from the desktop and running jobs from the command line.

Feedback and support

Important

Guest Users: For expired passwords and password resets, please contact RCS.

Our research computing environments are actively developed and continuously improving. Bug reports and feature requests are important contributions to this project and are always welcome and encouraged! If you find that something doesn't work as expected, of if you have a feature request, we want to know about it so we can fix or improve it.

There are several ways to connect:

  • For administrative requests including project and account creation or modification please use our request forms.

  • If you have a question, can't get something working, or if something looks broken you can reach out to us directly via email at research@hbs.edu.