Kappa Scheduling policy

Most Kappa nodes/projects use fair-share scheduling. Please see the Triolith page for details.

Differences between Kappa and Triolith scheduling:

  • Node sharing is not enabled on Kappa, all jobs will allocate at least one compute node.
  • Kappa has four development nodes. Use these by adding --reservation=devel -p devel to your sbatch/interactive options.
  • Kappa has 56 "fat" nodes with 72 GiB RAM.
  • Kappa has also 2 "huge" nodes with 1 TiB RAM. You need to apply for access to these nodes, please contact for more information.

AFM partition

26 Kappa nodes are owned solely by The Department of Physics, Chemistry and Biology within the center for "Advanced Functional Materials" (AFM) at Linköping University.

These are available to users who are members in the "afm" project (apply through NSC Express.

To use the AFM nodes, add "-p afm" to your sbatch/interactive options.

AFM scheduling is done manually within the AFM group by controlling who submits job. Ask the project PI Sergei Simak if you're unsure about this.

"Green-style" scheduling

92 Kappa nodes are available to the "liu1" and "liu5" projects. These use "Green-style" scheduling (named after the previous system "Green").

"Green-style" scheduling is very different from our normal fair-share scheduling. Users of this part of Kappa always have guaranteed access to a certain number of nodes. Since this setup can lead to a lot of wasted computing time when the guaranteed node are not used, "Green-type" scheduling also includes the possibility of running "risk jobs" to use that idle time. "Risk jobs" are jobs that might be killed at any time if a regular job needs to run. To avoid wasting computing time, you should make sure that all risk jobs use application-level checkpointing so that they can be restarted and continue where they left off. Risk jobs that are killed are not restarted by default. It is currently not possible to enable automatic restarting of killed jobs.

To submit jobs using "guaranteed nodes", add "-p green" to your sbatch/interactive options.

To submit risk jobs, add "-p green_risk" to your sbatch/interactive options.

Using the "huge" nodes

  • Apply for use of the huge nodes (Email support@nsc.liu.se and explain why you need them, and for how many core hours. Note that one hour of use "costs" 64 core hours regardless of how many CPU cores you actually use).
  • Add "-p huge -A $YOURPROJECT-huge" to your sbatch/interactive command line, e.g "interactive -N1 -p huge -t 8:00:00 -A liu-2014-00123-45-huge"