[OMPI users] bus error in openmpi-1.9a1r32669 on Solaris 10 Sparc

Hi, today I installed openmpi-1.9a1r32669 on my machines (Solaris 10 Sparc (tyr), Solaris 10 x86_64 (sunpc1), and openSUSE Linux 12.1 x86_64 (linpc1)) with Sun C 5.12 and gcc-4.9.0. I get the following bus error for my Sun C version on Solaris Sparc. tyr small_prog 125 ompi_info | grep -e MPI: -

[OMPI users] segmentation fault for Java in openmpi-1.9a1r32669 on Solaris 10 Sparc

Hi, today I installed openmpi-1.9a1r32669 on my machines (Solaris 10 Sparc (tyr), Solaris 10 x86_64 (sunpc1), and openSUSE Linux 12.1 x86_64 (linpc1)) with Sun C 5.12 and gcc-4.9.0. I get the following segmentation fault for my Sun C version on Solaris Sparc for Java programs. tyr java 137 ompi_

Re: [OMPI users] How does binding option affect network traffic?

I'm confused, then: why you wouldn't want to minimize the number of servers that a single job runs on? I ask because it sounds to me like you're running 12 jobs, each with 1 process per server. And therefore all 12 jobs are running on each server, like this: [cid:DE1F7C81-EFFB-45E3-8CDB-019B0A

Re: [OMPI users] How does binding option affect network traffic?

I am testing a new cluster that we just bought, which is why I am loading things this way. I am deliberately increasing network traffic. But in general, we submit jobs intermittently with various numbers of MPI processes. I have read that a good strategy is to map by socket, which in our case me

[OMPI users] enable-cuda with disable-dlopen

We found with 1.8.[1,2] that is you compile with --with-mxm --with-cuda --disable-dlopen OMPI will compile install and run, but if you run disabling mxm (say to debug something) mpirun --mca mtl ^mxm ./a.out You will get a notice saying that you cannot have enable cuda with disable dlopen,

Re: [OMPI users] enable-cuda with disable-dlopen

Yes, I have reproduced. And I agree with your thoughts on configuring vs runtime error. I will look into this. Thanks, Rolf >-Original Message- >From: users [mailto:users-boun...@open-mpi.org] On Behalf Of Brock Palen >Sent: Friday, September 05, 2014 5:22 PM >To: Open MPI Users >Subjec

[OMPI users] new overcommitment warning?

I suspect there is a new (to openmpi 1.8.N?) warning with respect to requesting a number of MPI processes greater than the number of "real" cores on a given machine. I can provide a good deal more information is that's required, but can I just pose it as a question for now? Does anyone know of

Re: [OMPI users] enable-cuda with disable-dlopen

Thanks Rolf. Sent from my iPhone > On Sep 5, 2014, at 6:28 PM, Rolf vandeVaart wrote: > > Yes, I have reproduced. And I agree with your thoughts on configuring vs > runtime error. I will look into this. > Thanks, > Rolf > >> -Original Message- >> From: users [mailto:users-boun...@o

Re: [OMPI users] new overcommitment warning?

On Sep 5, 2014, at 3:34 PM, Allin Cottrell wrote: > I suspect there is a new (to openmpi 1.8.N?) warning with respect to > requesting a number of MPI processes greater than the number of "real" cores > on a given machine. I can provide a good deal more information is that's > required, but ca

Re: [OMPI users] How does binding option affect network traffic?

On Sep 5, 2014, at 10:44 AM, McGrattan, Kevin B. Dr. wrote: > I am testing a new cluster that we just bought, which is why I am loading > things this way. I am deliberately increasing network traffic. But in > general, we submit jobs intermittently with various numbers of MPI processes. > I