[torquedev] [Bug 93] Resource management semantics of Torque need to be well defined

[Michel Béland]

Martin Siegert wrote:

> I do like the -lselect syntax in principle as long as can be
> introduced as an alternative to the -l nodes=x:ppn=y syntax.
> A few comments: I do not mind to use "ncpus", but others might:
> ncpus has a long history and I do not know whether anybody is still
> using it. In my opinion it is mostly broken, thus reusing "ncpus"
> for something else is not a big deal.

We do use -lncpus with Torque on Altix machines and although it is 
broken in some ways, -lnodes is too. I wrote a qsub wrapper that defines 
both -lncpus *and* -lnodes on Altix to make sure that everything would 
work correctly : routing based on processor count, routing based on 
properties, cpusets and displaying the number of tasks with qstat 
(although I think that the latter is fixed).

PBS Pro extended the meaning of ncpus and they define exactly what it 
means to use both -lncpus and -lselect=4:ncpus=8, although I do not 
think it is useful to specify both. Also, with a select request like 
above, it would define an attribute resources_used.ncpus=32 that would 
be used for routing. Very neat. Torque out to do the same...

> I do not like "mpiprocs" and "ompthread": there can be "procs" and
> "threads" other than "mpi" and "omp". We can use "threads" instead
> of "ompthread", but we cannot use "procs" instead of "mpiprocs" -
> that is taken already. Maybe we could use "nprocs" instead?

That is why I liked ppn and cpp: they are much shorter and were not tied 
to neither MPI nor OpenMP.

-- 
Michel Béland, analyste en calcul scientifique
michel.beland at rqchp.qc.ca
bureau S-250, pavillon Roger-Gaudry (principal), Université de Montréal
téléphone : 514 343-6111 poste 3892     télécopieur : 514 343-2155
RQCHP (Réseau québécois de calcul de haute performance)  www.rqchp.ca
Calcul Canada (computecanada.org)