Exactly, looks like your runtime estimation for your slot4 jobs is smaller than for your slot12 jobs. Backfilling must be active here. Did you submit both jobs in exactly the same way with s_rt? Try to set for your 4slot jobs the runtime higher than for your 12slot jobs.
Don't trust the default duration set in your configuration while other have runtime limits. There are some internal offsets handled for s_rt which are added. This could make the difference. So try to avoid to mix them up. Use one of the limits but for both. Daniel Am 25.05.2012 um 12:04 schrieb William Hay: > On 25 May 2012 10:31, Daniel Gruber <[email protected]> wrote: >> The expected behavior would be that when there is >> never an host with 12 slots free, that your cluster >> will be filled up with 4 slot jobs, even when they have >> lower priorities. The reservation you gave the 12 slot >> jobs will be attached at the end of that year. Hence > Doesn't that assume that the jobs on the nodes at present have the > default duration of a year > while the 4 slot jobs have something shorter? Otherwise the 12 slot > job would just keep getting pushed back. > > William _______________________________________________ users mailing list [email protected] https://gridengine.org/mailman/listinfo/users
