You might write a test of fork() and see if there's a Windows-related problem with that.  My guess is that this has nothing to do with MPICH and everything to do with the OS.  In any case, there's no point in discussing it further on the list unless you want to install Linux in a VM or something.

Jeff


On Mon, Aug 19, 2013 at 1:52 PM, Wesley Bland <wbland@mcs.anl.gov> wrote:
Unfortunately, we haven't supported Windows since MPICH version 1.4.1p1. You can find out more at the FAQ page: http://wiki.mpich.org/mpich/index.php/Frequently_Asked_Questions#Q:_Why_can.27t_I_build_MPICH_on_Windows_anymore.3F

Since we don't have any test environments or Windows experts, we can't really shed any light onto why you would get this performance. You might try MS-MPI as mentioned in the FAQ post.

Thanks,
Wesley

On Aug 19, 2013, at 12:44 PM, Abhishek Kumar Pandey <akpandey@adobe.com> wrote:

Hi,
 
I am using mpich-1.4.1.p1 to launch multiple child processes. On Window Server 2008 all the child processes are launched almost immediately but this takes significant time on Window Server 2012. 
 
I am using MPI_Comm_spawn for spawning a process and MPI::Init_thread(argc,argv,MPI_THREAD_MULTIPLE) is used for initializing MPICH2.
 
Please let me know if this is a known problem. Is there any workaround ? 
 
Regards,
Abhishek 
 
 
_______________________________________________
discuss mailing list     discuss@mpich.org
To manage subscription options or unsubscribe:
https://lists.mpich.org/mailman/listinfo/discuss


_______________________________________________
discuss mailing list     discuss@mpich.org
To manage subscription options or unsubscribe:
https://lists.mpich.org/mailman/listinfo/discuss



--
Jeff Hammond
jeff.science@gmail.com