Hi Gridengine users group members, I have the following problem - let me explain it with following example:
Node1 has 12 Cores/Slots and 1 MPI-Job with 2 Slots is running on it. A user submits job2 which require maximal 10 slots. Independently from schedule_interval, job_load_adjustments, load_formula and/or load_adjustment_decay_time parameters-settings job2 usually won't start on Node1 if 1) job2 doesn't start in the same schedule interval of job1, or 2) the user doesn't request exactly Node1 with the "-q" option, or 3) there are completely free nodes "The filling up mode" would make sense because we modified our scheduling with the tool "gepetools" from the GitHub website which allows requesting the number of nodes and slots per node similarly to PBS/Torque. It would be more efficient if "smaller" jobs share nodes per default and so reduce the waiting times of "bigger" jobs. So long I could not configure a satisfying solution (extending the scheduling interval has been the best option but also only works partly) and any new ideas are welcome. Ursula
_______________________________________________ users mailing list users@gridengine.org https://gridengine.org/mailman/listinfo/users
