Am 13.11.2010 um 15:39 schrieb Chris Jewell:
> Sorry for kicking off this thread, and then disappearing. I've been away for a bit. Anyway, Dave, I'm glad you experienced the same issue as I had with my installation of SGE 6.2u5 and OpenMPI with core binding -- namely that with 'qsub -pe openmpi 8 -binding set linear:1 <myscript.com>', if two or more of the parallel processes get scheduled to the same execution node, then the processes end up being bound to the same core. Not good!
> I've been playing around quite a bit trying to understand this issue, and ended up on the GE dev list:
as the [GE dev] list was nearly dead as it went closed source, I'm no longer subscripted to it. [GE users] will reach a broader audience I think. Anyway,
I don't have a free suitable cluster, but can you please try the following:
$ qsub -pe openmpi 8 -binding linear:2 <myscript.com>
with a fixed "allocation_rule 2" in your PE. And also:
$ qsub -pe openmpi 8 -binding linear:8 <myscript.com>
> It seems that most people expect that calls to 'qrsh -inherit' (that I assume OpenMPI uses to bind parallel processes to reserved GE slots) activates a separate binding. This does not appear to be the case. I *was* hoping that using -binding pe linear:1 might enable me to write a script that read the pe_hostfile and created a machine file for OpenMPI, but this fails as GE does not appear to give information as to which cores are unbound, only the number required.
You can get the information about the to be used cores when you use "env" or even better "pe" as "binding_instance" instead of "set". Then it should be possible (and you even need to implement it) to let Open MPI do the core binding instead of SGE. From `man qsub`:
pe means that the information about the selected cores appears in the fourth column of the pe_hostfile. Here the
logical core and socket numbers are printed (they start at 0 and have no holes) in colon separated pairs (i.e.
0,0:1,0 which means core 0 on socket 0 and core 0 on socket 1). For more information about the $pe_hostfile check
> So, for now, my solution has been to use a JSV to remove core binding for the MPI jobs (but retain it for serial and SMP jobs). Any more ideas??
> (PS. Dave: how is my alma mater these days??)
> Dr Chris Jewell
> Department of Statistics
> University of Warwick
> CV4 7AL
> Tel: +44 (0)24 7615 0778
> users mailing list