Page tree
Skip to end of metadata
Go to start of metadata

The Habanero High Performance Computing (HPC) Cluster is a service supporting research at Columbia.

All cluster inquiries should be directed to Please include the following information in your email:

  1. Your Columbia UNI.
  2. The name of the cluster you are working on ("Habanero" in this case).
  3. Job ID numbers, if asking about a job issue.

All computing at Columbia must abide by the Columbia University E-Mail & Computing Policies.

Habanero Back Online as of 6/18/2021

The Habanero cluster is back online and available for use. As mentioned in the previous communication, scratch storage data was lost on 5/16 due to a critical flaw in the metadata disk drives. The scratch storage system hardware that failed has been replaced and reset for use. Empty scratch spaces have been reconfigured for all groups and user home directory backups from mid-March have been restored.

Many of the most popular centrally installed software applications have been restored to /rigel/opt. However, the original environment module definition files were lost, so those are being recreated as needed. Due to the loss of some of this data, and the introduction of newer versions of certain software, job submission scripts may need to be updated to reflect changes. Some of these updates are reflected in the Habanero User Documentation. If you require software that was previously available on Habanero and is now missing, please let us know at

Getting Started

Submitting Jobs

Working on Habanero

Job Examples


Available software

Technical Information

Research Products

  • No labels