[torquedev] nodes, procs, tpn and ncpus

Glen Beane glen.beane at gmail.com
Wed Jun 9 19:53:46 MDT 2010


On Wed, Jun 9, 2010 at 9:01 PM, Garrick Staples <garrick at usc.edu> wrote:
> On Wed, Jun 09, 2010 at 08:52:08PM -0400, Glen Beane alleged:
>> On Wed, Jun 9, 2010 at 8:31 PM, Garrick Staples <garrick at usc.edu> wrote:
>> > I know I'm getting in on this conversation late, but here is my fantasy:
>> >
>> > nodes=X gives X number of cpus. Packed. Your job is CPU bound and you don't
>> > care how they are packed.
>>
>> blah.  that is overloading the meaning of nodes.  I like the new
>> procs=X instead. It basically means the same thing,  you get X
>> processors, moab seems to pack them on as few nodes as possible.
>> TORQUE doesn't do anything with procs yet...
>
> Nothing is overloaded. "nodes" has always translated to "vnodes" inside of
> torque. If you don't specify ppn, then you don't care about where your
> processors land. Perfectly logical. This case also covers the vast majority of
> jobs.
>
>
>> > nodes=X:ppn=Y gives you X unique nodes with Y cpus per machine. Not-packed.
>> > This lets you spread IO around because you know you need it.
>>
>>
>>
>> here is what I want
>>
>> procs=X gives you X processors, user doesn't care about layout (hack that
>> works with Moab, should be made to work properly with pbs_sched/qrun)
>> nodes=X:ppn=Y gives you exactly X unique nodes with Y processors per node
>> nodes=X - I'm not sure about this one, but to preserve historic behavior I
>> think TORQUE should give you X nodes with one processor on each node (Moab
>> can have an option to treat it like procs=X, which is the current behavior)
>> _______________________________________________ torquedev mailing list
>
> Getting torque to jive procs with nodes is a lot more work.
>
> My plan is easy, simple, and I think covers everyone's use cases.
>
> Everyone has always wanted "gimme X cores, anywhere". The solution is to not
> use EXACTNODE and "nodes=X" does what you want. But EXACTNODE breaks the
> "nodes=X:ppn=y" case. If we just change maui/moab to not pack jobs with ppn,
> then we are done.

Pretend Maui/Moab doesn't exist right now.  There is no way to do
"gimme X cores, anywhere" right?  What we're talking about is adding a
way to do that in TORQUE without scheduling hacks.

I'm already done because I use EXACTNODE and I use procs=X when I want
X processors and I don't care where they end up, and I use
nodes=X:ppn=Y if I do.  But it would be nice if TORQUE knew what to do
with procs.


More information about the torquedev mailing list