Getting started with HPC facility in TSD

Access Colossus

By default, to submit jobs on the Colossus cluster, users must log on to the submit host assigned to the project. You can reach this host by using PUTTY from your Windows VM, or by using ssh from your Linux VM in your TSD-project:

$ ssh pXX-hpc-01.tsd.usit.no

Data on the Colossus disk

When you get access to Colossus resources, you will have 1TiB of disk available on /cluster/projects/pXX (pXX = project number). In order to submit jobs you have to move all the input files and the data/repositories files needed in you workflow to the directory /cluster/projects/pXX using the submit host assigned to the project (as project login hosts can't access /cluster). This /cluster/projects/pXX directory is also where all your output files will be found at the end of the computation if the sbatch script is correctly written and the job completes successfully.