Re: [OMPI users] [EXTERNAL] Re: OpenMPI 4.0.5 error with Omni-path

2021-01-28 Thread Peter Kjellström via users
On Wed, 27 Jan 2021 15:31:40 -0500 Michael Di Domenico via users wrote: > if you have OPA cards, for openmpi you only need --with-ofi, you don't > need psm/psm2/verbs/ucx. I agree with Michael and would add for clarity that on the system you always need PSM2 and optionally libfabric (if you go t

Re: [OMPI users] MPI Exit Code:1 on an OpenFoam application

2021-01-28 Thread Kahnbein Kai via users
Tony, Im not able to run it in parallel on OF7. So i decide to use OF7 (with CfMesh) to mesh my case and then switch to OF8 and calculate it :). Thank you for your help and best regards Kai Am 10.01.21 um 16:49 schrieb Tony Ladd via users: Kai That means your case directory is mostly OK. E

Re: [OMPI users] [EXTERNAL] Re: OpenMPI 4.0.5 error with Omni-path

2021-01-28 Thread Heinz, Michael William via users
Patrick, A few more questions for you: 1. What version of IFS are you running? 2. Are you using CUDA cards by any chance? If so, what version of CUDA? -Original Message- From: Heinz, Michael William Sent: Wednesday, January 27, 2021 3:45 PM To: Open MPI Users Subject: RE: [OMPI users]

[OMPI users] Binding blocks of processes in round-robin fashion

2021-01-28 Thread Luis Cebamanos via users
Hello all, What are the options for binding MPI tasks on a blocks of cores per node/socket/numa in a round-robin fashion? Say I want to fully populate 40 core sockets on dual-socket nodes but in a round-robin fashion binding 4 cores on the first node, then 4 cores on the next, and so on.  Woul

Re: [OMPI users] Binding blocks of processes in round-robin fashion

2021-01-28 Thread Ralph Castain via users
Trying to wrap my head around this, so let me try a 2-node example. You want (each rank bound to a single core): ranks 0-3 to be mapped onto node1 ranks 4-7 to be mapped onto node2 ranks 8-11 to be mapped onto node1 ranks 12-15 to be mapped onto node2 etc.etc. Correct? > On Jan 28, 2021, at 3:0

Re: [OMPI users] Binding blocks of processes in round-robin fashion

2021-01-28 Thread Luis Cebamanos via users
That's right Ralph! On 28/01/2021 23:13, Ralph Castain via users wrote: Trying to wrap my head around this, so let me try a 2-node example. You want (each rank bound to a single core): ranks 0-3 to be mapped onto node1 ranks 4-7 to be mapped onto node2 ranks 8-11 to be mapped onto node1 ranks

Re: [OMPI users] Binding blocks of processes in round-robin fashion

2021-01-28 Thread Ralph Castain via users
Hmmm...well, the proc distribution is easy as you would just --map-by node. The tricky thing is assigning the ranks in the pattern you desire. We definitely don't have that pattern in our ranking algo today, though it wouldn't be hard to add. However, that wouldn't be available until OMPI v5 wa