Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.
Comment: added advice to srun to a compute node first before working.

Table of Contents

Getting Access

...

Account

Full Name

5sigma

Biostatistics

astroColumbia Astrophysics Lab
berkelbachChemistry
cboyceChemical Engineering
cklabIEOR
dbComputer Science
e3bEcology, Evolution and Environmental Biology
esmaSIPA-CGEP

exposomics

MSPH Exposomics
hillPhysics (Columbia Astrophysics Laboratory)
ieortangIndustrial Engineering and Operations Research (IEOR)
iicdIrving Institute for Cancer Dynamics 
intelseedfree

Special group with access to a non-NVIDIA GPU seed node from Intel. email hpc-support@columbia.edu if interested in details.

mcilvain

Grace McIlvain Lab 

morpheus

Bianca Dumitrascu Lab

msphMSPH IT

neuralctrl


Laboratory for Neural Engineering and Control
ntar_labBiomedical Engineering  (Morrison)
pas_labBiological Sciences
ueilBiomedical Engineering  (Konofagou)
qmechQuantum Mechanics/Applied Physics and Applied Math: Marianetti
ssccSocial Science Computing Committee (ISERP, Econ, and CPRC)
xulabEarth and Environmental Engineering


Your First Cluster Job

While best practice on all Columbia HPC group clusters, it is particularly important on Insomnia to move from the initial login node to a compute node before doing most work. Example:

Code Block
srun --pty -t 0-2:00 -A <ACCOUNT> /bin/bash

Now you have moved from the login node to a one of the compute nodes on the cluster. While simple things like editing a file or making new folders can be done on a login node, they can also be done on a compute node, and as you run more complicated jobs on Insomnia, some things simply will not work unless you are first on a compute node.

An example Submit Script


This script will print "Hello World", sleep for 10 seconds, and then print the time and date. The output will be written to a file in your current directory.

...