Hi,
Before starting programs on my cluster, I want to check on every CPU if it is
up and able to run MPI applications.
For this, I use a kind of 'ping' program that just send a message saying 'I'm
OK' tu a superviser program.
The 'ping' program is sent by the superviser on each CPU by the
MPI
Hi,
I have tried in 2 different clusters, and both times I have problems using the
headed node together with a headless one. There is no problem if I run 2
processes on n0, or 2 processes on n1, or on n2. No problem either using n1 and
n2. The problem is when I try to use n0 and n1, or n0 and
You're not doing anything wrong; it's just that Open MPI doesn't [yet]
handle failures well. It will probably *eventually* respond with a
timeout (and therefore fail).
You might want to run a real resource manager to manage your cluster,
such as SLURM, Torque, or one of a bunch of commercial solu
This problem still exists in OpenMPI 1.1a3r9704 (Apr 24, 2006), I
reported it for 9663 (Apr 20, 2006).
Michael
On Apr 21, 2006, at 12:32 AM, Jeff Squyres (jsquyres) wrote:
You're correct on all counts.
I've corrected the .h.sh script in the trunk and will get the correct
XSL (!) fixes in sh
Strange thing, with the latest g95 and the last OpenMPI 1.1 (a3r9704)
[on OS X 10.4.6] there does not seem to be the compilation penalty
for using "USE MPI" instead of "include mpi.h" that there used to be.
My test programs compile almost instantly.
However, I'm still seeing:
[a.b.c.d:2022
I apologize for the delay (and I actually do greatly appreciate your
reminders!). I made a change on the trunk back when I replied; I'm
waiting for my resident F90 expert to give me the "correct" fix (the
bindings are actually generated out of CHASM and XSL files) so that it
can be properly applie
Hi,
Looks like there is no problem with the setup of compute nodes. The error
seems to occur when trying to lookup the lower priority queue for a CQ
entry.
Thanks,
Aniruddha
- Original Message -
From: "Aniruddha Shet"
To: ; "Open MPI Users"
Sent: Monday, April 24, 2006 7:59 AM
Su
> -Original Message-
> From: users-boun...@open-mpi.org
> [mailto:users-boun...@open-mpi.org] On Behalf Of Michael Kluskens
> Sent: Tuesday, April 25, 2006 9:56 AM
> To: Open MPI Users
> Subject: [OMPI users] f90 module files compile a lot faster
>
> Strange thing, with the latest g95 and
I'm using OpenMPI 1.1 (today a3r9704) because 1.0.1 and 1.0.2 didn't
have enough support for the two MPI-2 commands I need for my project.
Because a tight deadline I'm not testing the trunk which is
openmpi-1.2a1r9704 at the moment.
Instead I'm making the fix to 1.1 I described below and t
If you're working in an SVN checkout, do this:
svn merge -r 9675:9676 https://svn.open-mpi.org/svn/ompi/trunk .
This will bring in what I did on the trunk to your checkout. Note that
it wasn't just the comm attr function that needed to be fixed, but also
the type and win functions, too.
>
Could I/we have a translation of what "trivial, small, medium, large"
means to the end user?
I for one don't read the docs every week with new 1.1 alpha tests.
Michael
On Apr 25, 2006, at 10:12 AM, Jeff Squyres (jsquyres) wrote:
-Original Message-
From: users-boun...@open-mpi.org
[m
> -Original Message-
> From: users-boun...@open-mpi.org
> [mailto:users-boun...@open-mpi.org] On Behalf Of Michael Kluskens
> Sent: Tuesday, April 25, 2006 11:15 AM
> To: Open MPI Users
> Subject: Re: [OMPI users] f90 module files compile a lot faster
>
> Could I/we have a translation of
I'm running OpenMPI 1.1 (v9704)and when a spawned processes exits the
parent does not die (see previous discussions about 1.0.1/1.0.2);
however, the next time the parent tries to spawn a process
MPI_Comm_spawn does not return.
My test output below:
parent: 0 of 1
parent: How many proce
Minor suggestion, change the first sentence to read:
- The Fortran 90 MPI bindings can now be built in one of four sizes
using --with-mpi-f90-size=SIZE.
Also, Open MPI 1.2 changes the --with-mpi-param-check default from
always to runtime according to my comparison of the 1.1 README and
1.
> -Original Message-
> From: users-boun...@open-mpi.org
> [mailto:users-boun...@open-mpi.org] On Behalf Of Michael Kluskens
> Sent: Tuesday, April 25, 2006 3:38 PM
> To: Open MPI Users
> Subject: Re: [OMPI users] f90 module files compile a lot faster
>
> Minor suggestion, change the first
15 matches
Mail list logo
