Paul, it would be incredibly helpful to reveal
* What version of Slurm you are using
* What Slurm commands you are using
* The mpirun command(s) that do effect what you desire
* Your slurm configuration -- preferably a copy of slurm.conf (with node
names and IP addresses obscured for security reasons)
* What OS you are using
Andy
On 11/13/2017 05:58 PM, Banks, Paul wrote:
Hi,
On my cluster I normally run LSP programs across multiple nodes with
mpirun (MVAPICH2) and can do that successfully, however I have always
had trouble getting it to run successfully with srun. Either it will
error out or the program will instead run multiple instances of the
same program across the nodes instead of sharing the workload. Now
this problem is impacting me when I try to set up an SBATCH routine.
Has anyone else run into a similar issue?
Best Regards,
Paul Banks
