Torque 2.3.0 cpusets broken [PATCH] (was Fwd: [torquedev] Quick test report of cpuset support)

Chris Samuel csamuel at vpac.org
Mon Apr 7 17:09:40 MDT 2008


----- "David Singleton" <David.Singleton at anu.edu.au> wrote:

> Yeah, I would suggest the right place for the subcpuset creation
> is in the application layer - best if MOM just creates the
> container cpuset.  Consider the case of a hybrid OpenMP/MPI app
> wanting two 4-thread tasks on each of your 8-core nodes.  That's
> a subcpuset configuration that might only be determinable by
> the application at runtime - couple climate codes can have
> nprocs/nthreads configuration in input files.

Mentioned it to Garrick in private email and he agrees that
making the per-vnode cpusets optional (and defaulting to off)
is a good plan.

He's tied up at his day job building a new cluster at the
moment though and so doesn't have time for Torque list email
for the moment.

cheers,
Chris
-- 
Christopher Samuel - (03) 9925 4751 - Systems Manager
 The Victorian Partnership for Advanced Computing
 P.O. Box 201, Carlton South, VIC 3053, Australia
VPAC is a not-for-profit Registered Research Agency


More information about the torquedev mailing list