Cirrus - Getting Started Guide

General information about the Cirrus cluster, including hardware specifications is availabe on the Cirrus system page.

Getting a login account

See the SMHI Access page and the MET Access page for more information.

How to log in

See the general instructions on getting started for information on how to login for the first time and set your permanent password.

• The Cirrus login nodes are cirrus1.nsc.liu.se and cirrus2.nsc.liu.se.

The SSH key fingerprints for both login nodes are:

SHA256:Sj538sFTdAs9ljYjZgA4ZKcket++q3een8ST+MkIMGw cirrus{1,2}.nsc.liu.se (ECDSA)
SHA256:gn0yHdi1zl6K7dnC/xUrJsZXTZSzCrPHLhBnXHlTGeg cirrus{1,2}.nsc.liu.se (ED25519)
SHA256:u7K/vU/EiLF8SWf0OCp3iS0qK6ie+J2M4OsLAyy2fUs cirrus{1,2}.nsc.liu.se (RSA)

About NSC’s cluster environment

Information about accessing the cluster and running jobs:

• Security
• Copying data
• Running applications
• Running graphical applications
• Batch jobs and scheduling (including Cirrus and Stratus specific information)
• Memory management

Storage

On Cirrus, NWP production data is stored in /nobackup/prod1, /nobackup/prod2, and /nobackup/fm. Group quota limits are in place for /nobackup/prod1 and /nobackup/prod2. Usage and quota information per group can be shown with the command nscquota -g, and per user with nscquota -a.

Available software and development tools

The software environment has been updated with respect to Alvin and Frost, to be more like NSC’s Tetralith and Sigma clusters for Swedish academic users. The cluster now runs CentOS 7 Linux, and the module system is slightly different: it is now based on Lmod, which has multiple hierarchies. NSC has also started installing software semi-automatically using the EasyBuild system (these modules are called [modulename]-eb). NSC strives to keep the software environment and the available software installations as identical as possible on all clusters.

• Compilers
• MPI libraries
• Math libraries
• All software pages