Raymond Wan <[email protected]> writes:

> RQS sounds like what I'm looking for (for #1) -- the problem I'm
> having is that users are being a bit greedy when I'm not watching.

It doesn't seem to have been said that RQS only affects scheduling the
job, not what happens at run time.  It specifically won't prevent a job
using more cores than it's asked for (assuming slots≡cores).

> Making all of the threads be on the same CPU is just a request from a
> user...it feels difficult to satisfy, but I'll look into it too.  I
> guess having this as an option (-binding) for qsub, as suggested by
> Reuti and William, for users to specify is ok.  I guess this is more
> like a user's wish list than a rule I'd like to enforce.

You really want core binding, but as Reuti says, how a multi-threaded
job uses the assigned cores is up to it, if it even chooses to obey it.
Typically you set something like GOMP_CPU_AFFINITY (for GNU OpenMP)
defined -- see past posts here.

-- 
Community Grid Engine:  http://arc.liv.ac.uk/SGE/

_______________________________________________
users mailing list
[email protected]
https://gridengine.org/mailman/listinfo/users

Reply via email to