On Wed, Dec 21, 2011 at 04:34:48AM -0500, Esztermann, Ansgar wrote:
Hello List,
is there some more detailed documentation available on how exactly share
tickets are distributed? I've found the "Configuring the share-based
policy" page on wikis.sun.com (seems to require a login now), but found
it lacking details.
There are a few sets of slides floating around online. I don't have the
link handy, but search for "dan templeton SGE course".
The problem at hand is this: sometimes, a user with no recorded usage
will submit jobs with very demanding resource requests, which consequently
No usage, no tickets assigned, or both?
sit around in qw for quite some time. They will also keep all available
share tickets to themselves, leaving none for other users' jobs. This is
OK for user A (no recorded usage means top priority), but users B and C
might have very different usage, and should therefore be assigned very
different priorities. However, there are no tickets left, so everyone
but user A is treated as an equal, upsetting fairness until those few
special nodes that can run A's jobs become free. Once A accumulates
usage, share tickets are gradually re-distributed to other users, and
things work fine; until user X becomes the new A, that is.
It sounds like you just may need more tickets. You didn't say how many
are in the total pool, but make sure it is fairly high--like 10,000.
Thanks,
A.
--
Ansgar Esztermann
DV-Systemadministration
Max-Planck-Institut für biophysikalische Chemie, Abteilung 105
_______________________________________________
users mailing list
[email protected]
https://gridengine.org/mailman/listinfo/users
--
Jesse Becker
NHGRI Linux support (Digicon Contractor)
_______________________________________________
users mailing list
[email protected]
https://gridengine.org/mailman/listinfo/users
