William Hay <[email protected]> writes:

>> my second problem, i have to find out the progress of reserving.
>> may be a job requests 4 slots, i have to find out the time when it
>> starts reserving, when it has got 2 slots, when it has got the 3rd
>> slot and so on.
>> 
> This appears to embody a slight misconception about reservations.  The
> reservations are for all the slots requested at some point in the
> future.  The slots may be used by other jobs up until that point in
> the future as backfill.  There is no point at which the reservation
> has only some of the requested resources.
>
> Potentially you could write a script to examine running jobs and work
> out when a given number of slots would be free if no new jobs started.
>  This might give you the information you want.

I don't think you could do it without duplicating the scheduler, except
in the simplest cases.  You need to take into account resource requests,
and access permissions, at least.  For example, jobs will run which
require site licence tokens is partly dictated by token use external to
the cluster, assuming you use something like flex-grid..

[For what it's worth the OAR DRM has "Gantt scheduling (so you can
visualize the internal scheduler decisions)", but I don't know what the
scheduling involves.]

-- 
Community Grid Engine:  http://arc.liv.ac.uk/SGE/
_______________________________________________
users mailing list
[email protected]
https://gridengine.org/mailman/listinfo/users

Reply via email to