We've successfully used IMPI 3.2.1 and 3.2.2 on our RHEL 4 cluster, but with 4.0.0 I get this:
Tell me please when should I install ICT on shared space and when on each node separetly. After tests I would like to create Linux Live CD to add nodes dynamicly, what kind of installation would be better for it? If you had any general advices in adding nodes dynamicly, I would be also very greatful :)
Thank you for help and regards.
I prepare network to Intel Cluster Toolkit installation. For test I get two computers.
- I called them master and node1.
- I added to /etc/hosts of both ip addresses.
- ping master and ping node1 commands works on both computers
- I created machines.LINUX file with lines:
My problem is when I run sshconnectivity.exp on master the only master's ssh settings are changed. No ~/.ssh directory appears on node1.
Thank you for help and regards
I am trying to run a program on a computer cluster. The structure of the program is the following:
We configure Torque on our cluster to set TMPDIR so that it points into a local file system on each node. A number of applications running on the cluster take advantage of TMPDIR. However, because the path is very long, TMPDIR causes mpd to fail. To avoid this problem, we set I_MPI_MPD_TMPDIR to /tmp. This works without a problem for Intel MPI 3.2.2, but with 4.0.0, a spurious error message appears on standard out stating that "Can't open file /tmp/mpd2.logfile...". For example, launching the test.c sample program with the mpirun command in a Torque script we get the following.
Tell me please, is Intel C++ compiler required for compiling applications that use Intel MPI or I can use some free compiler instead?
Hi I havesometimes problems with execution of a program with Intel MPI
It happens with an error on stderr (or stdout):
problem with execution of on wn20: [Errno 13] Permission denied
What could be a problem?
here is my ulimit -a:
Does anyone knoe ifthere isa way to define a IB PKEY and Service_Level for intel MPI?
For open MPI you can set the env var:
I get the following error on my cluster when I submit jobs
mpiexec_node050: cannot connect to local mpd (/tmp/mpd2.console_sudharshan); possible causes:
1. no mpd is running on this host
2. an mpd is running but was started without a "console" (-n option)
I'm trying to build mpich2 on an SGI Altix (256-processor IA-64 platform), using the Intel compilers (c/c++ and fortran). If I run configure with gcc, all works fine; the configure completes to the end successfully. However, if I run it specifying the intel compilers, it fails with an error on missing atomic primitive availability...no.