OMPI always tries to use the lowest numbered address first - just a natural
ordering. You need to tell it to use just the public ones for this topology.
Use the oob_tcp and btl_tcp parameters to do this. See "ompi_info --param oob
tcp" and "ompi_info --param btl tcp" for the exact syntax.
Sent
We are constructing a set of computers with Open MPI and there's a small
problem with mixing public and private IPs.
We aren't sure about what's causing the problem or how to solve it.
The files are shared thanks to NFS and we have a couple computers with
private IPs and public IPs that we want t
Looks like a bug - can address next week. Very unusual use of npernode...
Sent from my iPad
On Oct 4, 2011, at 4:55 AM, Andrew Senin wrote:
> Hi all,
>
> I noticed a strange behaviour in 1.5.4 which seems to me as a bug. I'm trying
> to launch 4 ranks on 2 nodes. If I add "-npernode 3 -byno
Hi all,
I noticed a strange behaviour in 1.5.4 which seems to me as a bug. I'm
trying to launch 4 ranks on 2 nodes. If I add "-npernode 3 -bynode"
parameters mpirun freezes on start. When I tried with the "-display-map"
parameter it shows the following:
./mpirun --np 4 --host mir1,mir2 -npernode 3
Hello people,
I want to use MPI-Get asynchronously, is it possible
? The scenario is like, the process will initiate an MPI-Get call
asynchronously for some data from neighbouring processes and will continue with
its work and it may check at the end of performing the
