Hi, Am 28.02.2014 um 07:15 schrieb Riccardo Murri:
> Hi Reuti, > > On 27 February 2014 13:41, Reuti <[email protected]> wrote: >>> So we have limits configured in GE that should restrain any user from >>> using more than 32 cores in the very-short/interactive queue: >>> >>> murri@login2:~> qconf -srqs >>> { >>> name pe_in_very-short.q >>> description "Limit parallel jobs on very-short.q to 32 slots max." >>> enabled TRUE >>> limit users {*} queues {very-short.q} to slots=32 >> >> limit users {*} queues very-short.q to slots=32 >> >> should do too. > > The man page `sge_resource_quota` seems to make a distinction between > using braces or not, but I cannot make sense of the relevant paragraph: > > By default, the expressed limit counts for the entire filter scope. To > express a filter-scope-specific limit, it’s possible to define an > expanded list by setting the list between ’{’ ’}’. > > What's the difference between limits that "count for the entire filter > scope" and "filter-scope-specific limits"? With curly braces: limit applies to each of the listed entries on its own Without curly brcaes: limit applies to the sum across all listed entries If there is just one entry, it's the same of course. But disregarding {} may ease the computation. >>> However, parallel jobs requesting h_rt <= 1800 can be allocated a mix >>> of "very-short" queue instances and slots from other queues. This is >>> a sample of such jobs: >>> >>> 2703080: very-short.q@r08c01b08n01 >>> 2703080: very-short.q@r08c01b07n01 >>> 2703080: short.q@r08c02b07n02 >>> 2703080: short.q@r06c01b09n02 >>> >>> 2703081: very-short.q@r08c01b07n02 >>> 2703081: very-short.q@r08c01b11n01 >>> 2703081: short.q@r08c01b02n02 >>> 2703081: very-short.q@r08c01b11n02 >>> >>> 2703093: short.q@r06c01b09n01 >>> 2703093: short.q@r01c02b09n01 >>> 2703093: long.q@r07c03b10n02 >>> 2703093: very-short.q@r08c01b12n01 >>> >>> So it is still possible for a user to consume all the `very-short` >>> slots by submitting many jobs. In other words, the RQS shown above is >>> applied *per job* and not *per user*. >> >> How many slots were used in each line in the above example - the violation >> of 32 is not obvious? > > Ah, sorry - every queue instance counts as 8 slots. Has the PE a fixed allocation rule of 8 or how can this be determined? -- Reuti >> You can force a job to stay in one queue only by creating a PE for >> each queue (and attach it only to this queue) and submit with a >> wildcard in the PE request: > > Thanks for the suggestion. But it's actually fine if jobs cross > queues, it's just that no user should be able to use more than 4 nodes > in the `very-short.q` at any given time... > > Thanks, > Riccardo > -- > Riccardo Murri > http://www.gc3.uzh.ch/people/rm > > Grid Computing Competence Centre > University of Zurich > Winterthurerstrasse 190, CH-8057 Zürich (Switzerland) > Tel: +41 44 635 4222 > Fax: +41 44 635 6888 > _______________________________________________ users mailing list [email protected] https://gridengine.org/mailman/listinfo/users
