Re: [OMPI users] OpenMPI 1.6.3 and Memory Issues

2012-11-29 Thread Joseph Farran
Hi again. I am using /etc/modprobe.d/mofed.conf, otherwise I get: WARNING: Deprecated config file /etc/modprobe.conf, all config files belong into /etc/modprobe.d/ But I am still getting the memory errors after making the changes and rebooting: $ cat /etc/modprobe.d/mofed.conf options mlx4_co

Re: [OMPI users] OpenMPI 1.6.3 and Memory Issues

2012-11-29 Thread Yevgeny Kliteynik
You can also set these parameters in /etc/modprobe.conf: options mlx4_core log_num_mtt=24 log_mtts_per_seg=1 -- YK On 11/30/2012 2:12 AM, Yevgeny Kliteynik wrote: > On 11/30/2012 12:47 AM, Joseph Farran wrote: >> I'll assume: /etc/modprobe.d/mlx4_en.conf > > Add these to /etc/modprobe.d/mofed

Re: [OMPI users] cluster with iOS or Android devices?

2012-11-29 Thread Beatty, Daniel D CIV NAVAIR, 474300D
Greetings Ladies and gentlemen, There is one alternative approach and this a psuedo-cloud based MPI. The idea is that MPI node list is adjusted via the cloud similar to the way Xgrid's Bonjour used to do it for Xgrid. In this case, it is applying an MPI notion to the OpenCL codelets. There are o

Re: [OMPI users] OpenMPI 1.6.3 and Memory Issues

2012-11-29 Thread Yevgeny Kliteynik
On 11/30/2012 12:47 AM, Joseph Farran wrote: > I'll assume: /etc/modprobe.d/mlx4_en.conf Add these to /etc/modprobe.d/mofed.conf: options mlx4_core log_num_mtt=24 options mlx4_core log_mtts_per_seg=1 And then restart the driver. You need to do it on all the machines. -- YK > > On 11/29/2012 0

Re: [OMPI users] CentOS 6.3 & OpenMPI 1.6.3

2012-11-29 Thread Joseph Farran
Hi YK: Yes, I have those installed but they are newer versions: # rpm -qa | grep rdma librdmacm-1.0.15-1.x86_64 librdmacm-utils-1.0.15-1.x86_64 librdmacm-devel-1.0.15-1.x86_64 # locate librdmacm.la # Here are the RPMs the Mellanox build created for kernel: 2.6.32-279.14.1.el6.x86_64 # ls *rdm

Re: [OMPI users] CentOS 6.3 & OpenMPI 1.6.3

2012-11-29 Thread Yevgeny Kliteynik
Joseph, On 11/29/2012 11:50 PM, Joseph Farran wrote: > make[2]: Entering directory > `/data/apps/sources/openmpi-1.6.3/ompi/mca/mtl/mxm' > CC mtl_mxm.lo > CC mtl_mxm_cancel.lo > CC mtl_mxm_component.lo > CC mtl_mxm_endpoint.lo > CC mtl_mxm_probe.lo > CC mtl_mxm_recv.lo > CC mtl_mxm_send.lo > CCLD

Re: [OMPI users] OpenMPI 1.6.3 and Memory Issues

2012-11-29 Thread Joseph Farran
I'll assume: /etc/modprobe.d/mlx4_en.conf On 11/29/2012 02:34 PM, Joseph Farran wrote: Where do change those mellanox settings? On 11/29/2012 02:23 PM, Jeff Squyres wrote: See http://www.open-mpi.org/faq/?category=openfabrics#ib-low-reg-mem. On Nov 29, 2012, at 5:21 PM, Joseph Farran wrote:

Re: [OMPI users] OpenMPI 1.6.3 and Memory Issues

2012-11-29 Thread Joseph Farran
Where do change those mellanox settings? On 11/29/2012 02:23 PM, Jeff Squyres wrote: See http://www.open-mpi.org/faq/?category=openfabrics#ib-low-reg-mem. On Nov 29, 2012, at 5:21 PM, Joseph Farran wrote:

Re: [OMPI users] OpenMPI 1.6.3 and Memory Issues

2012-11-29 Thread Jeff Squyres
See http://www.open-mpi.org/faq/?category=openfabrics#ib-low-reg-mem. On Nov 29, 2012, at 5:21 PM, Joseph Farran wrote: > Hi All. > > In compiling a simple Hello world with OpenMPI 1.6.3 and mpirun the hello > program, I am getting: > > $ ulimit -l unlimited > $ mpirun -np 2 hello > --

[OMPI users] OpenMPI 1.6.3 and Memory Issues

2012-11-29 Thread Joseph Farran
Hi All. In compiling a simple Hello world with OpenMPI 1.6.3 and mpirun the hello program, I am getting: $ ulimit -l unlimited $ mpirun -np 2 hello -- WARNING: It appears that your OpenFabrics subsystem is configured to onl

Re: [OMPI users] CentOS 6.3 & OpenMPI 1.6.3

2012-11-29 Thread Joseph Farran
On 11/28/2012 10:53 AM, Mike Dubman wrote: You need mxm-1.1.3a5e745-1.x86_64-rhel6u3.rpm On Wed, Nov 28, 2012 at 7:44 PM, Joseph Farran mailto:jfar...@uci.edu>> wrote: mxm-1.1.3a5e745-1.x86_64-rhel6u3.rpm After installing MLNX_OFED_LINUX-1.5.3-3.1.0-rhel6.3-x86_64, removing the old mx

[OMPI users] MPI_Allreduce with F90 Handles

2012-11-29 Thread Shane Hart
All, I have a Fortran code that works quite well with OpenMPI 1.4.3 where I create a handle using: call MPI_TYPE_CREATE_F90_INTEGER(9, COMM_INT4, ierror) and then do a reduction with: call MPI_ALLREDUCE(send_buffer, buffer, count, COMM_INT4, MPI_SUM, communicator, ierror) Howev

Re: [OMPI users] error with mrmpi and openmpi

2012-11-29 Thread Ralph Castain
Ah, thanks - I was curious as Greenplum is about to release the full port of Hadoop to OpenMPI so it can be run anywhere and support MPI as well. I'm at least a little familiar with this one, but didn't realize it had been distributed. On Nov 29, 2012, at 8:00 AM, Howard Pritchard wrote: > H

Re: [OMPI users] Can not turn off C++ build.

2012-11-29 Thread Jeff Squyres
No problem! Glad to help. I added you to the ticket about not being able to turn off the C++ compiler checks (https://svn.open-mpi.org/trac/ompi/ticket/2999), in case that ever gets fixed. It's somewhat of a low priority. On Nov 29, 2012, at 11:17 AM, Ray Sheppard wrote: > Thanks Jeff, > O

Re: [OMPI users] Can not turn off C++ build.

2012-11-29 Thread Ray Sheppard
Thanks Jeff, Of course you were right. I had thought the lost function was something internal to y'alls build. It is pretty scary that they have been building and porting for weeks (while I was running around SC and the holidays) and it takes an old fortran guy to notice they don't have a wo

Re: [OMPI users] error with mrmpi and openmpi

2012-11-29 Thread Howard Pritchard
Hi Ralph, mrmpi is an mpi based map reduce implementation developed at sandia labs. Howard On 11/28/2012 09:20 PM, Ralph Castain wrote: On Nov 28, 2012, at 12:21 PM, Mariana Vargas Magana wrote: Hi openmpi'users I now trying to install mrmpi in a cluster to use it with openmpi, I install

Re: [OMPI users] error with mrmpi and openmpi

2012-11-29 Thread Jeff Squyres
On Nov 28, 2012, at 11:20 PM, Ralph Castain wrote: >> libibverbs: Warning: no userspace device-specific driver found for >> /sys/class/infiniband_verbs/uverbs0 > > Looks like OMPI was built with Infiniband support, but we aren't finding the > required support libraries wherever the process is r

[OMPI users] mca_btl_tcp_frag_recv: readv failed: Connection timed out (110)

2012-11-29 Thread Syed Ahsan Ali
I recieve following error while running an application Does this represent any hardware issue? [compute-01-01.private.dns.zone][[60090,1],10][btl_tcp_frag.c:216:mca_btl_tcp_frag_recv] mca_btl_tcp_frag_recv: readv failed: Connection timed out (110) [compute-01-01.private.dns.zone][[60090,1],13][btl