The Galaxy gpuq/workq are not technically not for our use (though practically we can still use the gpuq for a while). The two alternatives are to use the workq on either Zeus or Magnus. In order to do this we need the script templates and/or wrapper scripts to allow the cluster to be an option.
Ideas:
default to Zeus
default to current cluster if script is run from either magnus/zeus login node
use -M as a cluster flag as this is inline with the slurm directive and not yet used by any of our wrapper scripts
The Galaxy gpuq/workq are not technically not for our use (though practically we can still use the gpuq for a while). The two alternatives are to use the workq on either Zeus or Magnus. In order to do this we need the script templates and/or wrapper scripts to allow the cluster to be an option.
Ideas:
-M
as a cluster flag as this is inline with the slurm directive and not yet used by any of our wrapper scripts