Il 07/06/20 09:44, Diego Zuccato ha scritto: >> I'm *guessing* that you are tripping over the use of "--tasks 32" on a >> heterogeneous cluster, > If you mean that using "--tasks 32" trips the use of a second node, then > no. The node does have two AMD Opteron 6274 . [...] > I've had a similar problem while adding new nodes in a new partition. I > "solved" (probably) by adding a line > mtl = psm2 > to /etc/openmpi/openmpi-mca-params.conf . > But those were nodes with IB. Update: it's not resolved on these nodes either. :(
I have another partition on these new nodes. 4 identical machines, new installation, ConnectX-5 card, dual Intel Xeon 5120 (14 core dual thread). No problem running a job requiring 112 threads (on 4 nodes), but can't run a single-node job with 56 threads. -- Diego Zuccato DIFA - Dip. di Fisica e Astronomia Servizi Informatici Alma Mater Studiorum - Università di Bologna V.le Berti-Pichat 6/2 - 40127 Bologna - Italy tel.: +39 051 20 95786
