<div dir="ltr">people, shmem, posix or not is technology from yesterday ! wake up.</div><div class="gmail_extra"><br><br><div class="gmail_quote">2013/6/3 Rolf Rabenseifner <span dir="ltr"><<a href="mailto:rabenseifner@hlrs.de" target="_blank">rabenseifner@hlrs.de</a>></span><br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Dear implementers of MPI-3.0 MPI_WIN_ALLOCATE_SHARED,<br>
<br>
my question is not about the Interface, but on the implementations:<br>
<br>
If a set of MPI processes within a shared memory node and Communicator<br>
wants to allocate a shared memory window:<br>
<br>
- Are there similar limits as if a set of threads wants to allocate<br>
memory and use it as global Memory by all threads, i.e.,<br>
is it possible to allocate most physical<br>
memory with MPI_WIN_ALLOCATE_SHARED?<br>
<br>
- Is there an additional restriction on the number of shared memory<br>
windows?<br>
<br>
- Is there an additional restriction if one process defines the<br>
whole size as window size and all other processes within<br>
the SMP node use size=0?<br>
<br>
- Do you know other restrictions if I want to substitute<br>
OpenMP (not OpenMPI!) by shared memory MPI programming,<br>
i.e. hybrid MPI+MPI instead of hybrid MPI+OpenMP?<br>
<br>
Is your answer generally valid for mpich and OpenMPI and must<br>
I expect some additional restrictions on some vendor's MPIs?<br>
<br>
It is about the real implementations, e.g. in mpich and OpenMPI,<br>
but I expect that our MPI-3 RMA working group knows the answer.<br>
<br>
Best regards<br>
<span class="HOEnZb"><font color="#888888">Rolf<br>
<br>
<br>
--<br>
Dr. Rolf Rabenseifner . . . . . . . . . .. email <a href="mailto:rabenseifner@hlrs.de">rabenseifner@hlrs.de</a><br>
High Performance Computing Center (HLRS) . phone <a href="tel:%2B%2B49%280%29711%2F685-65530" value="+4971168565530">++49(0)711/685-65530</a><br>
University of Stuttgart . . . . . . . . .. fax <a href="tel:%2B%2B49%280%29711%20%2F%20685-65832" value="+4971168565832">++49(0)711 / 685-65832</a><br>
Head of Dpmt Parallel Computing . . . <a href="http://www.hlrs.de/people/rabenseifner" target="_blank">www.hlrs.de/people/rabenseifner</a><br>
Nobelstr. 19, D-70550 Stuttgart, Germany . . . . (Office: Room 1.307)<br>
_______________________________________________<br>
mpi3-rma mailing list<br>
<a href="mailto:mpi3-rma@lists.mpi-forum.org">mpi3-rma@lists.mpi-forum.org</a><br>
<a href="http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-rma" target="_blank">http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-rma</a><br>
</font></span></blockquote></div><br></div>