[OMPI users] MX MPI latency measurements

With 2 nodes using 1.1.7 with the patch we measured (using mpich-mx 1.2.7..4): 3.07us With mx 1.2.1-rc18 we measure: 3.69 us And with mpich-mx 1.2.7..4 we see: 3.20us Our Open MPI settings: --- # env | grep OMPI OMPI_MCA_pml=cm OMPI_MCA_mpi_keep_hostnames=1 OMPI_MCA_oob_tc

Re: [OMPI users] Intel MPI SendRcv Performance using MX 1.1.7 comparison of Open MPI 1.2.2 and MPICH-MX 1.2.7..4

rt the results? > > Thanks, > > Galen > > > On May 18, 2007, at 1:15 PM, Maestas, Christopher Daniel wrote: > > > Hello, > > > > I was wondering why we would see ~ 100MB/s difference > between mpich-mx > > and Open MPI with SendRecv from the

[OMPI users] Intel MPI SendRcv Performance using MX 1.1.7 comparison of Open MPI 1.2.2 and MPICH-MX 1.2.7..4

Hello, I was wondering why we would see ~ 100MB/s difference between mpich-mx and Open MPI with SendRecv from the Intel MPI benchmarks. Maybe I'm missing turning something on? The hardware is: --- # mx_info -q MX Version: 1.1.7 MX Build: root@tocc1:/projects/global/SOURCES/myricom/mx-1.1.7 Fri M

Re: [OMPI users] Pernode request

M > To: Open MPI Users > Subject: Re: [OMPI users] Pernode request > > > > > On 12/12/06 9:18 AM, "Maestas, Christopher Daniel" > > wrote: > > > Ralph, > > > > I figured I should of run an mpi program ...here's what it > do

Re: [OMPI users] Pernode request

t specify "byslot", then we default > to assigning > ranks by node. > > Make sense? If so, I can probably have that going before the holiday. > > Ralph > > > > On 12/11/06 7:51 PM, "Maestas, Christopher Daniel" > > wrote: > >

[OMPI users] Multiple mpiexec's within a job (schedule within a scheduled machinefile/job allocation)

Hello, Sometimes we have users that like to do from within a single job (think schedule within an job scheduler allocation): "mpiexec -n X myprog" "mpiexec -n Y myprog2" Does mpiexec within Open MPI keep track of the node list it is using if it binds to a particular scheduler? For

Re: [OMPI users] Pernode request

Hello Ralph, This is great news! Thanks for doing this. I will try and get around to it soon before the holiday break. The allocation scheme always seems to get to me. From what you describe that is how I would have seen it. As I've gotten to know osc mpiexec through the years I think they li

Re: [OMPI users] Pernode request

Ralph, I agree with what you stated in points 1-4. That is what we are looking for. I understand your point now about the non-MPI users too. :-) Thanks, -cdm -Original Message- From: users-boun...@open-mpi.org [mailto:users-boun...@open-mpi.org] On Behalf Of Ralph Castain Sent: Wednesda

Re: [OMPI users] Pernode request

Ralph, Thanks for the feedback. Glad we are clearing these things up. :-) So here's what osc mpiexec is doing now: --- -pernode : allocate only one process per compute node -npernode : allocate no more than processes per compute node --- > Cdm> I think I originally requested the -pernode

Re: [OMPI users] Pernode request

Thanks for the feedback Ralph. Comments below. -Original Message- From: users-boun...@open-mpi.org [mailto:users-boun...@open-mpi.org] On Behalf Of Ralph Castain Sent: Tuesday, November 28, 2006 3:27 PM To: Open MPI Users Subject: Re: [OMPI users] Pernode request We already have --perno

[OMPI users] Pernode request

I recently saw this on the mpiexec mailing list and pondered that this would be a useful feature for Open MPI as well. :-) I can't seem to enter a trac ticket and seem to be having issues w/ my browser at the moment, but wanted to get this out there. --- > > > 1) mpiexec already has "-pernode" but

[OMPI users] Quote on mvapich site

I believe the quote regarding thunderbird on the following site is not correct: http://nowlab.cse.ohio-state.edu/projects/mpi-iba/ We do have mvapich installed on thunderbird, but I believe the quote is misleading in leading people to believe mvapich was used to obtain our recent top500 nu

Re: [OMPI users] OMPI launching problem using TM and openib on 1920 nodes

Some more background information: 1) the environment is all run inside an initrd with a static pbs_mom. 2) the file we change in the torque distributions is: torque-2.1.2/src/include/dis.h --- 255 /* NOTE: increase THE_BUF_SIZE to 131072 for systems > 5k nodes */ 256 257 /* OLD: #define T

[OMPI users] Question on mpi collectives

How fast/well are MPI collectives implemented in ompi? I'm running the Intel MPI 1.1. benchmarks and seeing the need to set wall clock times > 12 hours for run sizes of 200 and 300 nodes for 1ppn and 2ppn cases. The collective tests that usually pass in 2ppn cases: Barrier, Reduce scatter, allred

[OMPI users] Orte_error_log w/ ompi 1.1.1 and torque 2.1.2

Has anyone ever seen this? --- [dn32:07156] [0,0,0] ORTE_ERROR_LOG: Temporarily out of resource in file base/rmaps_base_node.c at line 153 [dn32:07156] [0,0,0] ORTE_ERROR_LOG: Temporarily out of resource in file rmaps_rr.c at line 270 [dn32:07156] [0,0,0] ORTE_ERROR_LOG: Temporarily out of resource

[OMPI users] A -pernode behavior using torque and openmpi

Hello, I was wondering if openmpi had a -pernode like behavior similar to osc mpiexec mpiexec -pernode mpi_hello Would launch N mpi processes on N nodes ... No more no less. Openmpi already will try and run N*2 nodes if you don't specify -np mpirun -np mpi_hello Launches N*2