Skip to Main Content
HBS Home
  • About
  • Academic Programs
  • Alumni
  • Faculty & Research
  • Baker Library
  • Giving
  • Harvard Business Review
  • Initiatives
  • News
  • Recruit
  • Map / Directions
Research Computing Services
  • Online Requests
  • FAQ
  • Blog
  • Contact Us
  • About Us
  • Faculty Projects
  • Training
  • Compute Cluster & Data Storage
  • Data Practices
  • Help
  • …→
  • Harvard Business School→
  • Research Computing Services→
  • Compute Cluster and Data Storage
    • Compute Cluster and Data Storage
    • Compute Cluster
    →
  • Compute Cluster
    • Compute Cluster
    • Running Jobs
    • Software Tools
    →
  • Running Jobs
    • Running Jobs
    • Guidelines for Choosing Resources
    • Running a Program/Submitting a Job
    • Monitoring Jobs & Troubleshooting
    • Scaling Work
    →

Compute Cluster

Compute Cluster

  • Running Jobs
  • Software Tools

Running Jobs

Running Jobs

  • Compute Cluster
    • Running Jobs
      • Guidelines for Choosing Resources
      • Running a Program/Submitting a Job
      • Monitoring Jobs & Troubleshooting
      • Scaling Work
    • Software Tools
7ms
The most important goal for using the HBSGrid is for running programs and scripts that might consume more resources than are available on your desktop or laptop. These sections will provide information to give you a complete understanding of the modes for using the compute cluster as well as details for doing so effectively.

Guidelines for Choosing Resources 

The HBSGrid's collection of compute nodes and servers are all coordinated by the software system Load Sharing Facility, otherwise known as LSF (see LSF reference guide), or the scheduler. This scheduler listens to your requests and then matches you with the computing resources that you need to perform your work. These resources are exclusively available to you for the duration of your job, but is part of a larger shared system with finite constraints, so it is important to be judicious in your resource usage. The section on choosing your resources provides guidelines for doing this.

Running a Program/Submitting a Job 

There are two sets of methods of running programs and jobs on the HBS compute Grid: using default submission (wrapper) scripts or by writing custom submission commands. Default submission (wrapper) scripts are the most straightforward way to submit a job on the HBSGrid. However, with this ease of use, there is the tradeoff in flexibility: there are limited choices in what programs can be run and the resources requested.

  • Interactive sessions are appropriate for doing work in graphical applications, interactive "console tools" like R and iPython, significant script/code development, or efforts like iterative data exploration, either in a GUI or at the command line. For example, signing in via the NoMachine client, and starting an application from the Applications drop-down menus is common (but not the only) way start an interactive session.
  • Batch jobs are sessions or program runs that are executed in the background and where there isn't any user interaction. This is useful for finished scripts or code you wish to run repeatedly, such as parameter sweeps or large analyses broken into smaller parts (pleasantly parallelizing). This can be done through the command line or via the Program Application Center. Scripts written for graphical applications like MATLAB, Stata, and SAS, can be run in this manner and scaled for large numbers of runs, unleashing the power of the compute cluster. Using the Program Application Center to start an application with a run script is an example of running a batch job.

Monitoring Jobs & Troubleshoting 

Monitoring the progress of your jobs and understanding how to control them is essential. This section provides key commands to accomplish this, along with guidance on troubleshooting.

Scaling Work 

Learn how to scale up your work through such techniques such as parallel processing.

ǁ
Campus Map
Research Computing Services (RCS) 
Harvard Business School
Baker Library, B90, 25 Harvard Way
Boston, MA 02163
Phone: 617.495.6100
Email: research@hbs.edu
→Map & Directions
→More Contact Information
→Terms Of Service
  • Make a Gift
  • Site Map
  • Jobs
  • Harvard University
  • Trademarks
  • Policies
  • Accessibility
  • Digital Accessibility
Copyright © President & Fellows of Harvard College.