Hi folks
I'm trying to run an MPI app on an infiniband cluster with OpenMPI
1.2.6.
When run on a single node, this app is grabbing large chunks of memory
(total per process ~8.5GB, including strace showing a single 4GB grab)
but not using it. The resident memory use is ~40MB per process. When
t
If OMPI is spinning consuming 100% of your CPU, it usually means that
some MPI function call is polling waiting for completion. Given the
pattern you are seeing, I'm wondering if some Open MPI collective call
is not finishing until you re-enter the MPI progression engine.
Specifically, is
On May 11, 2008, at 12:00 AM, Wen Hao Wang wrote:
1. How can I change/set Fortran compiler used after installing
openmpi?
I installed openmpi-1.2.5-2 on RHEL5.2 Server with x86_32 arch. The
rpm has been shipped RHEL images. I directly installed the rpms
without compiling the source code. No
On May 10, 2008, at 2:57 PM, Alberto Giannetti wrote:
Please don't CC both mailing lists on future replies to this thread;
one or the other would be fine; thanks!
Brad Benton and George Bosilca are the release managers for the v1.3
series. They're maintaining a wiki for the v1.3 series here:
Hi,
I think that ubuntu's package version might have some problem. I just
downloaded source code from open mpi and it just worked after source
installation.
Thank you,
Sang Chul
P.S. The code was as follows. I ran it with $ mpirun -n 2 hello
Hi,
I'm using Openmpi in a linux cluster (itanium 64, intel compilers, 8
processors (4 dual) by node) in which openmpi is not the default ( I
mean supported) MPI-II implementation. Openmpi has been installed easily
on the cluster but I think there is a problem with the configuration.
I'm usi
ello,
I have Rocks 4.3 installed. Frontend 32bit and nodes with 64bit.
Traying to run openmpi with torque support using multilib.
/usr/local shared among nodes
CONFIGURING and INSTALL
[from frontend 32bit]
./configure --with-tm=/opt/torque/ --prefix=/usr/local/openmpi-1.2.6 --
includedir='${pr
