On 11/16/2010 08:24 PM, Ralph Castain wrote:
I do not think you are right here. Chris sent the following which
looks like OGE (fka SGE) actually did bind the hnp to multiple
cores. However that message I believe is not coming from the
processes themselves and actually is only shown by the hnp. I
wonder if Chris adds a "-bind-to-core" option we'll see more output
from the a.out's before they exec unterm?
On Tue, Nov 16, 2010 at 12:23 PM, Terry
On 11/16/2010 01:31 PM, Reuti wrote:
Am 16.11.2010 um 15:40 schrieb Ralph Castain:
2. have SGE bind procs it launches to -all- of those cores. I believe SGE does this automatically to constrain the procs to running on only those cores.
This is another "bug/feature" in SGE: it's a matter of discussion, whether the shepherd should get exactly one core (in case you use more than one `qrsh`per node) for each call, or *all* cores assigned (which we need right now, as the processes in Open MPI will be forks of orte daemon). About such a situtation I filled an issue a long time ago and "limit_to_one_qrsh_per_host yes/no" in the PE definition would do (this setting should then also change the core allocation of the master process):
I believe this is indeed the crux of the issue
fantastic to share the same view.
FWIW, I think I agree too.
3. tell OMPI to --bind-to-core.
In other words, tell SGE to allocate a certain number of cores on each node, but to bind each proc to all of them (i.e., don't bind a proc to a specific core). I'm pretty sure that is a standard SGE option today (at least, I know it used to be). I don't believe any patch or devel work is required (to either SGE or OMPI).
When you use a fixed allocation_rule and a matching -binding request it will work today. But any other case won't be distributed in the correct way.
Is it possible to not include the -binding request? If SGE is told to use a fixed allocation_rule, and to allocate (for example) 2 cores/node, then won't the orted see
itself bound to two specific cores on each node?
When you leave out the -binding, all jobs are allowed to run on any core.
We would then be okay as the spawned children of orted would inherit its binding. Just don't tell mpirun to bind the processes and the threads of those MPI procs will be able to operate across the provided cores.
Or does SGE only allocate 2 cores/node in that case (i.e., allocate, but no -binding given), but doesn't bind the orted to any two specific cores? If so, then that would be a problem as the orted would think itself unconstrained. If I understand the thread correctly, you're saying that this is what happens today - true?
Exactly. It won't apply any binding at all and orted would think of being unlimited. I.e. limited only by the number of slots it should use thereon.
So I guess the question I have for Ralph. I thought, and
this might be mixing some of the ideas Jeff and I've been
talking about, that when a RM executes the orted with a
bound set of resources (ie cores) that orted would bind the
individual processes on a subset of the bounded resources.
Is this not really the case for 1.4.X branch? I believe it
is the case for the trunk based on Jeff's refactoring.
You are absolutely correct, Terry, and the 1.4 release
series does include the proper code. The point here, though,
is that SGE binds the orted to a single core, even though
other cores are also allocated. So the orted detects an
external binding of one core, and binds all its children to
that same core.
Sure. Here's the stderr of a job submitted to my cluster with 'qsub -pe mpi 8 -binding linear:2 myScript.com' where myScript.com runs 'mpirun -mca ras_gridengine_verbose 100 --report-bindings ./unterm':
[exec4:17384] System has detected external process binding to cores 0022
[exec4:17384] ras:gridengine: JOB_ID: 59352
[exec4:17384] ras:gridengine: PE_HOSTFILE: /usr/sge/default/spool/exec4/active_jobs/59352.1/pe_hostfile
[exec4:17384] ras:gridengine: exec4.cluster.stats.local: PE_HOSTFILE shows slots=2
[exec4:17384] ras:gridengine: exec2.cluster.stats.local: PE_HOSTFILE shows slots=1
[exec4:17384] ras:gridengine: exec7.cluster.stats.local: PE_HOSTFILE shows slots=1
[exec4:17384] ras:gridengine: exec3.cluster.stats.local: PE_HOSTFILE shows slots=1
[exec4:17384] ras:gridengine: exec6.cluster.stats.local: PE_HOSTFILE shows slots=1
[exec4:17384] ras:gridengine: exec1.cluster.stats.local: PE_HOSTFILE shows slots=1
[exec4:17384] ras:gridengine: exec5.cluster.stats.local: PE_HOSTFILE shows slots=1
What I had suggested to Reuti was to not include the
-binding flag to SGE in the hopes that SGE would then bind the
orted to all the allocated cores. However, as I feared, SGE in
that case doesn't bind the orted at all - and so we assume the
entire node is available for our use.
This is an SGE issue. We need them to bind the orted to
-all- the allocated cores (and only those cores) in order for
us to operate correctly.
users mailing list
Terry D. Dontje | Principal Software Engineer
Engineering | +1.781.442.2631
Oracle - Performance
95 Network Drive,
Burlington, MA 01803