Il 19/06/2016 18:15, Achi Hamza ha scritto: > mpirun -np 4 lmp_openmpi < in.vacf.2d Are you sure you're running the job on the worker nodes and not on the head node? Happens all the time, here :) Try confirming with a parallel htop (maybe via cssh) on all worker nodes. Or slurmtop.
If the job is actually running on the nodes, then the network could be the bottleneck (expecially if the home is NFS-mounted and uses the same network interface MPI is using). HIH -- Diego Zuccato Servizi Informatici Dip. di Fisica e Astronomia (DIFA) - Università di Bologna V.le Berti-Pichat 6/2 - 40127 Bologna - Italy tel.: +39 051 20 95786 mail: diego.zucc...@unibo.it 5x1000 AI GIOVANI RICERCATORI DELL'UNIVERSITÀ DI BOLOGNA Codice Fiscale: 80007010376
