Hi, I just saw this. 

We definitely need to consider this concern.  

I also need to go review the APIs.

Thanks,
Tony


On Sat, Jul 14, 2018 at 12:27 AM, Junchao Zhang <jczhang@mcs.anl.gov> wrote:
 I want to try MPI neighborhood collectives. I have a communication graph and want to do both forward and backward communication along the edges. With current APIs, it looks I need to create two comm_dist_graphs. It is a waste since MPI implementations can do similar optimizations for the two.
 Should MPI support this scenario?  Thanks.

--Junchao Zhang

_______________________________________________
mpiwg-coll mailing list
mpiwg-coll@lists.mpi-forum.org
https://lists.mpi-forum.org/mailman/listinfo/mpiwg-coll




--
Tony Skjellum, PhD
RunTime Computing Solutions, LLC
tony@runtimecomputing.com
direct: +1-205-918-7514
cell: +1-205-807-4968