Hi Diego
I (still) have Torque/PBS version 4.something in old clusters.
[Most people at this point already switched to Slurm.]
Torque/PBS comes with a tool named "pbsdsh" (for PBS distributed shell):
http://docs.adaptivecomputing.com/torque/4-1-3/Content/topics/commands/pbsdsh.htm
t,
or were not relayed to the list subscribers.
Thank you,
Gus Correa
> Begin forwarded message:
>
> From: Gustavo Correa
> Subject: Re: [OMPI users] MPI advantages over PBS
> Date: August 25, 2018 at 14:41:36 EDT
> To: Open MPI Users
>
> Hi Diego
>
> I (still) ha
The message may not matter,
but somehow two copies of this sent earlier today didn't make it to the list.
Gus Correa
> Begin forwarded message:
>
> From: Gustavo Correa
> Subject: Re: [OMPI users] MPI advantages over PBS
> Date: August 25, 2018 at 14:41:36 EDT
> To: Open
Somehow two copies of this sent earlier today didn't make it to the list.
Gus Correa
> Begin forwarded message:
>
> From: Gustavo Correa
> Subject: Re: [OMPI users] MPI advantages over PBS
> Date: August 25, 2018 at 20:16:49 EDT
> To: Open MPI Users
>
> Hi Diego
&g
Hi Diego
I (still) have Torque/PBS version 4.something in old clusters.
[Most people at this point already switched to Slurm.]
Torque/PBS comes with a tool named "pbsdsh" (for PBS distributed shell):
http://docs.adaptivecomputing.com/torque/4-1-3/Content/topics/commands/pbsdsh.htm
Hi Diego
I (still) have Torque/PBS version 4.something in old clusters.
[Most people at this point already switched to Slurm.]
Torque/PBS comes with a tool named "pbsdsh" (for PBS distributed shell):
http://docs.adaptivecomputing.com/torque/4-1-3/Content/topics/commands/pbsdsh.htm
Hi Max
Name resolution in /etc/hosts is a simple solution for (2).
I hope this helps,
Gus
> On May 15, 2018, at 01:39, Max Mellette wrote:
>
> Thanks everyone for all your assistance. The problem seems to be resolved
> now, although I'm not entirely sure why these changes
On Jun 5, 2016, at 12:03 PM, Alexander Droste wrote:
> Hi everyone,
>
> I'd like to know what the maximum buffer size
> for sends/receives is. Besides the count being limited
> to INT_MAX, how is the max buffer size limited?
>
> Best regards,
> Alex
>
Hi Alexander
As far as I know, the
Hi Ehsan
You didn't tell the details of how you configured and installed Open MPI.
However, you must point the configuration --prefix to the installation
directory, say:
./configure --prefix=/opt/openmpi-1.8.8
In addition, the installation directory must be *different* from the
directory where
Thank you, George!
Sorry that I sent my message too quick, and didn't even write that the
bug report was from MPICH (although the link was to MPICH).
Anyway, glad to know that it helped.
So, for the time being (OMPI 1.8.4 and 1.6.5) is it safer to stick to
MPI_Type_Vector,
instead of venturing
Besides Gilles suggestions, wouldn't this:
> TYPE tParticle
> INTEGER :: ip
INTEGER :: dummy
DOUBLE PRECISION :: RP(2)
DOUBLE PRECISION :: QQ(2)
>
> ENDTYPE tParticle
(instead of making RP and QQ real)
match better your declarations of (MPI) TYPES()
Hi Dave, Gilles, list
There is a problem with knem in OMPI 1.8.3.
A fix is supposed to come on OMPI 1.8.4.
Please, see this long thread:
http://www.open-mpi.org/community/lists/users/2014/10/25511.php
Note also, as documented in the thread,
that in the OMPI 1.8 series "vader" replaces "sm" as
Hi Lee-Ping
Did you cleanup the old build, to start fresh?
make distclean
configure --disable-vt ...
...
I hope this helps,
Gus Correa
On Sep 29, 2014, at 8:47 AM, Lee-Ping Wang wrote:
> Hmm, the build doesn't finish - it breaks when trying to create the man page.
> I guess I'll disable
Hi Elie
Glad to know you solved the problem. Congrats!
Hopefully our discussion will help other EPW+QE users
that subscribe to the Open MPI list troubleshoot EPW problems.
Indeed, the Quantum Espresso (QE) configure script seems to miss the correct
library names
of the Intel MKL libraries
Hi Alain
Have you tried this?
mpiexec -np 2 ./a.out
Note: mpicc to compile, mpiexec to execute.
I hope this helps,
Gus Correa
On May 26, 2014, at 9:59 AM, Alain Miniussi wrote:
>
> Hi,
>
> I have a failure with the following minimalistic testcase:
> $: more ./test.c
> #include "mpi.h"
>
>
I guess you need to provide buffers of char type to
MPI_Send and MPI_Recv, not NULL.
On Mar 16, 2014, at 8:04 PM, Elias Rudberg wrote:
> Hi Ralph,
>
> Thanks for the quick answer!
>
>> Try running the "ring" program in our example directory and see if that works
>
> I just did this, and it
FWIW, I never added any verbs or openib switches to the OMPI configure command
line, and configure found them and built with Infiniband support.
Up to OMPI 1.6.5 (I didn't try the 1.7 series).
Unless you are installing your ofed/IB packages in non-standard places,
I guess configure will find
On Feb 14, 2014, at 5:59 AM, Reuti wrote:
> Am 14.02.2014 um 11:23 schrieb tmish...@jcity.maeda.co.jp:
>
>> You've found it in the dream, interesting!
>
> It happens sometimes to get insights while dreaming:
>
>
Is your IB card in compute-01-10.private.dns.zone working?
Did you check it with ibstat?
Do you have a dual port IB card in compute-01-15.private.dns.zone?
Did you connect both ports to the same switch on the same subnet?
TCP "no route to host":
If it is not a firewall problem, could it bad
O_MAIL=/var/spool/mail/leeping
>> PBS_NUM_PPN=12
>> PBS_NODEFILE=/opt/torque/aux//272139.certainty.stanford.edu
>> PBS_O_PATH=/opt/intel/Compiler/11.1/064/bin/intel64:/opt/intel/Compile
>> r/11.1
>> /064/bin/intel64:/usr/local/cuda/bin:/home/leeping/opt/psi-4.0b5/bin:/
>> home/l
he batch system
> as effectively as possible within the confines of their policies. This is
> why it makes sense to submit a single multiple-node batch job, which then
> executes several independent single-node tasks.
>
> - Lee-Ping
>
> -Original Message-
> From: user
l.
>
> As for the queue policy, that is a highly political issue because the
> cluster is a shared resource. My usual recourse is to use the batch system
> as effectively as possible within the confines of their policies. This is
> why it makes sense to submit a single multiple-n
t; I will try your suggestions and get back to you. Thanks!
>
> - Lee-Ping
>
> -Original Message-
> From: users [mailto:users-boun...@open-mpi.org] On Behalf Of Gustavo Correa
> Sent: Saturday, August 10, 2013 12:04 PM
> To: Open MPI Users
> Subject: Re: [OMPI users] Er
?
2) Look for the -machinefile switch in their script $QC/bin/parallel.csh
and replace it by
-hostfile $PBS_NODEFILE.
My two cents,
Gus Correa
On Aug 10, 2013, at 3:03 PM, Gustavo Correa wrote:
> Hi Lee-Ping
>
> I know nothing about Q-Chem, but I was confused by these sentences:
>
> &qu
Hi Lee-Ping
I know nothing about Q-Chem, but I was confused by these sentences:
"That is to say, these tasks are intended to use OpenMPI parallelism on each
node, but no parallelism across nodes. "
"I do not observe this error when submitting single-node jobs."
"Since my jobs are only
Hi Jed
You could try to select only ethernet interface that match your node's IP
addresses,
which seems to be en2.
The en1 interface seems to be an external IP.
Not sure about en3, but it is awkward that it has a
different IP than en2, but in the same subnet.
I wonder if this may be the
Hi Derek
Since your mpif90 is installed in /opt/local/lib/openmpi/bin/mpif90
shouldn't the openmpi libraries be located in /opt/local/lib/openmpi/lib,
rather than /opt/local/lib that your "--showme" shows?
I wonder if this is an environment variable problem
(PATH and
Dear Open MPI Pros
Somehow I am stuck offsite and I have to test/develop an MPI program on a super
duper
2006 vintage Mac PowerBookPro with Mac OS X 10.6.8 (Snow Leopard).
This is a 32-bit machine with dual core Intel Core Duo processors and 2GB RAM.
Well, my under-development program using
On Mar 30, 2013, at 10:02 AM, Duke Nguyen wrote:
> On 3/30/13 8:20 PM, Reuti wrote:
>> Am 30.03.2013 um 13:26 schrieb Tim Prince:
>>
>>> On 03/30/2013 06:36 AM, Duke Nguyen wrote:
On 3/30/13 5:22 PM, Duke Nguyen wrote:
> On 3/30/13 3:13 PM, Patrick Bégou wrote:
>> I do not know
Hi Tetsuya Mishima
In your example, have you tried not to modify the node file,
launch two mpi processes with mpiexec, and request a "-bynode" distribution of
processes:
mpiexec -bynode -np 2 ./my_program
This should launch one MPI process in each of two nodes.
See 'man mpiexec' for details.
On Nov 24, 2012, at 11:38 PM, Diego Avesani wrote:
> dear all,
> thanks for the web site.
>
> About the previous question. Can I compile with only ifort?
> I mean:
> udo ./configure --prefix=/opt/ompi-1.6 FC=ifort
> avoiding the others?
Hi
This will probably build openMPI with the native C/C++
On Mar 16, 2012, at 8:51 AM, Addepalli, Srirangam V wrote:
> This usually means you library path is not updated to find mpilibrarues. You
> can fix this many ways, basic two steps are
>
> 1. Identify location of your libraries (use locate, find )
> 2. Add it to your Library path. ( export
Hi Joshua
Can you your int counter "i" get so large?
> for(i=0; i<=1; i++)
I may be mistaken, but
1,000,000,000,000 = 10**12 > 2**31=2,147,483,647=maximum int.
Unless they are 64-bit long[s].
Just a thought.
Gus Correa
On Mar 13, 2012, at 4:54 PM, Joshua Baker-LePain wrote:
On Mar 6, 2012, at 3:59 PM, Kharche, Sanjay wrote:
>
> Hi
>
> I am working on a 3D ADI solver for the heat equation. I have implemented it
> as serial. Would anybody be able to indicate the best and more
> straightforward way to parallelise it. Apologies if this is going to the
> wrong
Hi Timothy
There is no call to MPI_Finalize in the program.
Would this be the problem?
I hope this helps,
Gus Correa
On Mar 6, 2012, at 10:19 AM, Timothy Stitt wrote:
> Will definitely try that. Thanks for the suggestion.
>
> Basically, I need to be able to scatter values from a sender to a
Hi Claudio
Check 'man mpirun'.
You will find examples of the
'-byslot', '-bynode', '-loadbalance', and rankfile options,
which allow some control of how ranks are mapped into processors/cores.
I hope this helps,
Gus Correa
On Mar 1, 2012, at 2:34 PM, Claudio Pastorino wrote:
> Hi, thanks
ff you don't want. Use pgcc, pgcc, or icc directly.
>>>>> You
>>>>> shouldn't have any trouble running in parallel with either aprun or
>>>>> mpirun
>>>>> (orterun).
>>>>>
>>>>> If you haven't already please check
GI -Bstatic flag, perhaps?
This type of thing only the developers can tell.
Better hear from them.
I hope this helps,
Gus Correa
> Abhinav
>
> On Wed, Feb 15, 2012 at 7:42 AM, Gustavo Correa <g...@ldeo.columbia.edu>
> wrote:
>> Hi Abhinav
>>
>> Did you add to y
.
>
> I guess I can safely ignore these warnings and run on Ethernet, but it would
> be nice to know what happened there, in case anybody has an idea.
>
> Thank you,
>
> On Wed, Feb 15, 2012 at 12:52 AM, Gustavo Correa <g...@ldeo.columbia.edu>
> wrote:
> Hi
n I am still getting the "attempted
> static link of dynamic object" error as I reported earlier.
>
> Thanks.
>
> On Fri, Feb 10, 2012 at 5:57 AM, Gustavo Correa <g...@ldeo.columbia.edu>
> wrote:
>> Hi Abhinav
>>
>> Setting CC='pgcc
Hi Tohiko
OpenFabrics network a.k.a. Infiniband a.k.a. IB.
To check if the compute nodes have IB interfaces, try:
lspci [and search the output for Infinband]
To see if the IB interface is configured try:
ifconfig -a [and search the output for ib0, ib1, or similar]
To check if the OFED module
gt;> Error: unknown option "--daemonize"
>> even though I am just running with -np 2 -hostfile test.hst
>>
>> The program runs fine on 2 cores if running locally on each node.
>>
>> Any ideas ??
>>
>> Thanks
>>
>> Richard
>>
On Feb 13, 2012, at 11:02 AM, Richard Bardwell wrote:
> Ralph
>
> I had done a make clean in the 1.2.8 directory if that is what you meant ?
> Or do I need to do something else ?
>
> I appreciate your help on this by the way ;-)
Hi Richard
You can install in a different directory, totally
Hi Abhinav
Setting CC='pgcc --noswitcherror', FC='pgfortran --noswitcherror', etc, may
prevent the
error message [but not sure it will prevent any actual and unreported error].
Check details with 'man pgfortran'.
I hope this helps,
Gus Corea
.
On Feb 10, 2012, at 2:03 AM, Abhinav Sarje
Hi Frank, Lloyd
If all you want is to sort out from which process the output is coming from,
you can use the "-tag-output" switch to the [OpenMPI] mpirun.
Check it out with 'man mpirun'.
I hope this helps,
Gus Correa
On Feb 1, 2012, at 11:04 AM, Lloyd Brown wrote:
> I don't know about using
.0.0127.0.0.1 127.0.0.1 1
> 255.255.255.255 255.255.255.255 255.255.255.255 2 1
> ===
> Persistent Routes:
> None
>
> -Original Message-
> From: use
Have you tried to specify the hosts with something like this?
mpirun -np 2 -host localhost ./my_program
See 'man mpirun' for more details.
I hope it helps,
Gus Correa
On Jan 16, 2012, at 6:34 PM, MM wrote:
> hi,
>
> when my wireless adapter is down on my laptop, only localhost is
Hi Mark
I wonder if you need to initialize the module command environment inside your
SGE
bash submission script:
$MODULESHOME/init/
where is bash in this case. See 'man module' for more details.
This would be before you actually invoke the module command:
module load openmpi
I am
Hi Richard
I have 1.4.4 built with Intel 12.0. It works.
Any chance that your Intel-based OpenMPI was built from a source
directory that had been previously used to build the PGI-based OpenMPI,
and no 'make distclean' was issued in between the two builds,
nor a fresh build done from a brand
I don't know about the grid engine/ SGE.
However, in Torque, the batch/resource manager I use,
to allow oversubscription, you need to modify the batch server nodes file
and pretend the nodes have more cores than the physical ones.
[Something like 'node01 np=8' would change to 'node01 np=16' for
rrectly in the environment.
>
> On Dec 21, 2011, at 1:08 PM, Gustavo Correa wrote:
>
>> You probably need also to launch the program with mpiexec (mpiexec -np 4
>> ./hello_c),
>> not just ./hello_c as your email suggests you did.
>>
>> On Dec 21, 2011, at 2
You probably need also to launch the program with mpiexec (mpiexec -np 4
./hello_c),
not just ./hello_c as your email suggests you did.
On Dec 21, 2011, at 2:12 PM, Ralph Castain wrote:
> Did you remember to set your LD_LIBRARY_PATH to include /opt/openmpi, per
> your configure line?
>
>
>
emporary arrays than if you just pass
> the correct subarray. Anyway, this is not really an MPI issue as much as an
> F90 bug in your compiler.
>
> On 12/14/11 08:57, Gustavo Correa wrote:
>> Hi Patrick
>>
>> > From my mere MPI and Fortran-90 user point of vi
, Component v1.4.2)
> MCA ess: env (MCA v2.0, API v2.0, Component v1.4.2)
> MCA ess: hnp (MCA v2.0, API v2.0, Component v1.4.2)
> MCA ess: singleton (MCA v2.0, API v2.0, Component v1.4.2)
> MCA ess: slurm (MCA v2.0, API v2.0, Compone
of things to look for?
>
> Thank you,
>
> On Wed, Dec 14, 2011 at 11:05 AM, Gustavo Correa <g...@ldeo.columbia.edu>
> wrote:
> Hi Micah
>
> Is ifort in your PATH?
> If not, the OpenMPI configure script will use any fortran compiler it finds
> first, w
Hi Patrick
>From my mere MPI and Fortran-90 user point of view,
I think that the solution offered by the MPI standard [at least up to MPI-2]
to address the problem of non-contiguous memory layouts is to use MPI
user-defined types,
as I pointed out in my previous email.
I like this solution
Hi Micah
Is ifort in your PATH?
If not, the OpenMPI configure script will use any fortran compiler it finds
first, which may be gfortran.
You need to run the Intel compiler startup script before you run the OpenMPI
configure.
The easy thing to do is to source the Intel script inside your
On Dec 12, 2011, at 8:42 AM, amjad ali wrote:
> Thanking you all very much for the reply.
>
> I would request to have some reference about what Tim Prince & Andreas has
> said.
>
> Tim said that OpenMPI has had effective shared memory message passing. Is
> that anything to do with
Hi Patrick
I think tab(i,:) is not contiguous in memory, but has a stride of nbcpus.
Since the MPI type you are passing is just the barebones MPI_INTEGER,
MPI_BCAST expects the four integers to be contiguous in memory, I guess.
The MPI calls don't have any idea of the Fortran90 memory layout,
and
able-mem-profile
>
> You don't need --with-devel-headers unless you intend to write code that
> directly drives the OMPI internals.
>
>
> On Dec 7, 2011, at 10:00 AM, John Doe wrote:
>
>> Hi Gustavo,
>>
>> I do have /opt/ompi/gnu/1.4.4/lib in my LD_LIBRARY_PATH and
>> directory in my path as well but that didn't seem to help.
>>
>> Sam
>>
>> On Tue, Dec 6, 2011 at 5:18 PM, Gustavo Correa <g...@ldeo.columbia.edu>
>> wrote:
>> Hi John Doe
>>
>> What you need to add to LD_LIBRARY_PATH
Hi John Doe
What you need to add to LD_LIBRARY_PATH is /opt/ompi/gnu/1.4.4/lib
[note 'lib' at the end].
Your email seems to say that you added /opt/ompi/gnu/1.4.4/lib/openmpi
instead, if I understood it right.
And to your PATH you need to add the corresponding 'bin' directory:
rt to v1.4 from the original Trac ticket
> https://svn.open-mpi.org/trac/ompi/ticket/2586.
>
> It *looks* like this patch is supposed to allow -tp to pass through
> Libtool.
>
> The Libtool sh here is quite complex; I'm not sure I'm reading it right. Can
> you try -tp=shan
> Hope that helps,
>
> Samuel K. Gutierrez
> Los Alamos National Laboratory
>
> On Nov 8, 2011, at 2:06 PM, Gustavo Correa wrote:
>
>> Dear OpenMPI pros
>>
>> When I try to build OpenMPI 1.4.4 with PGI compilers 11.7 [pgcc, pgcc,
>> pgfortran]
>>
Dear OpenMPI pros
When I try to build OpenMPI 1.4.4 with PGI compilers 11.7 [pgcc, pgcc,
pgfortran]
I get the awkward error message on the bottom of this email.
I say awkward because I assigned the value 'shanghai-64' to the '-tp' flag,
as you can see from the successful 'libtool:compile'
Hi Ashwani
Also, check if there are rogue processes from old jobs in your
compute nodes taking up lots of file descriptors.
A reboot should fix this easily.
My two cents.
Gus Correa
On Oct 15, 2011, at 10:34 AM, Ralph Castain wrote:
> Okay, let's try spreading them out more, just to avoid
Hi Amos
This form perhaps?
'export PATH=/opt/openmpi/bin:$PATH'
You don't want to wipe off the existing path, just add openmpi to it.
Also, Intel also has its shared libraries, which may be causing trouble.
My guess is that you need to set the Intel environment first by
placing a line more or
Hi Tena
Thank you for taking the time to explain the details of
the EC2 procedure.
I am afraid everything in my bag of tricks was used.
As Ralph and Jeff suggested, this seems to be a very specific
problem with EC2.
The difference in behavior when you run as root vs. when you
run as Tena, tells
do
> > this using openmpi.
>
> You probably have to start by defining what you mean by "the time it
> took for the communication". Anyhow, the Peruse instrumentation in OMPI
> might help.
>
>
> --
>
> Message: 5
>
On Feb 1, 2011, at 5:02 PM, Jeffrey A Cummings wrote:
> I use OpenMPI on a variety of platforms: stand-alone servers running Solaris
> on sparc boxes and Linux (mostly CentOS) on AMD/Intel boxes, also Linux
> (again CentOS) on large clusters of AMD/Intel boxes. These platforms all
> have
On Feb 1, 2011, at 1:09 AM, Bibrak Qamar wrote:
> Hello All,
>
> I am using non-blocking send and receive, and i want to calculate the time it
> took for the communication. Is there any method or a way to do this using
> openmpi.
>
> Thanks
> Bibrak Qamar
> Undergraduate Student BIT-9
>
Hi Brock
He's probably hitting the MPI address boundary of 2GB.
A workaround is to declare a user defined type (MPI_TYPE_CONTIGUOUS,
or MPI_TYPE_VECTOR), to bundle a bunch of primitive data (e.g. reals),
then send (broadcast for him/her) a smaller number of those types.
See this thread:
s annoying. <323.gif>
>
> Regards,
> Benjamin
>
> 2010/12/6 Gustavo Correa <g...@ldeo.columbia.edu>
> Hi Benjamin
>
> I would just rebuild OpenMPI withOUT the compiler flags that change the
> standard
> sizes of "int" and "float"
Hi Daofeng
It is hard to tell what is happening in the Infiniband side of the problem.
Did somebody perhaps remove the Infiniband card from this machine?
Was it ever there?
Did somebody perhaps changed the Linux kernel modules that are loaded
(perhaps changing /etc/module.config or similar)?
Hi Benjamin
I would just rebuild OpenMPI withOUT the compiler flags that change the standard
sizes of "int" and "float" (do a "make cleandist" first!), then recompile your
program,
and see how it goes.
I don't think you are gaining anything by trying to change the standard
"int/integer" and
Hi Jack
Your code snippet is too terse, doesn't show the MPI calls.
It is hard to guess what is the problem this way.
Gus Correa
On Oct 24, 2010, at 5:43 PM, Jack Bryan wrote:
> Thanks for the reply.
> But, I use mpi_waitall() to make sure that all MPI communications have been
> done before
Hi Jack
It may depend on "do some things".
Does it involve MPI communication?
Also, why not put MPI_Finalize();return 0 outside the ifs?
Gus Correa
On Oct 24, 2010, at 2:23 PM, Jack Bryan wrote:
> Hi
>
> I got a problem of open MPI.
>
> My program has 5 processes.
>
> All of them can
Hi Brandon
Yes, as others said and you figured, version is important,
and installing from source is not hard, probably will save you more headaches
than
installing the packages (which normally vary with the Linux distro version).
Regarding the environment variables.
You need to set the PATH
Hi Brandon
You must have passwordless ssh setup across the machines.
Check if you can ssh passwordless back and forth across all node pairs,
with the host names or IPs you have in your host.txt file.
Your /etc/host (or whatever Ubuntu uses to match hosts and IPs) must be
consistent (perhaps the
Check PETSc:
http://www.mcs.anl.gov/petsc/petsc-as/
On Jul 18, 2010, at 12:37 AM, Damien wrote:
> You should check out the MUMPS parallel linear solver.
>
> Damien
> Sent from my iPhone
>
> On 2010-07-17, at 5:16 PM, Daniel Janzon wrote:
>
>> Dear OpenMPI Users,
>>
>> I
80 matches
Mail list logo