[ofiwg] strawman of the datatypes to be supported

Jeff Hammond jeff.science at gmail.com
Tue Sep 9 10:42:21 PDT 2014

o AR: Jeff Hammond to provide a strawman of the datatypes to be supported

I propose that the following MPI datatypes have explicit support in
the API.  As Chapter 4 of MPI-3 is extremely detailed in its
definition of these operations, I see no reason to summarize here.  I
only list the motivation and some caveats.

The over-arching caveat is that I do not think we need to support
recursive type creating the way MPI does.  The derived types below
only need to be composed of built-in types.


General purpose IOVEC supports allows the user to act on arbitrary
noncontiguous data using a single API call.

2) MPI_TYPE_VECTOR (shmem_<T>_i{put,get} are special cases of this
where blocklength=1)

Vector types are an obvious optimization because communicating such
data regions otherwise requires O(n) operations or a single IOVEC
operation with O(n) metadata, where n is the number of elements in the
vector.  Explicit expression of vector types also enables the use of
scatter-gather engines in the CPU and/or NIC.


This is an optimization to (1) that reduces the vector of block sizes
to a scalar.

This is not necessarily a complete list of type support but my three
highest priorities as a user and/or implementor of MPI-3, GA/ARMCI and



On Tue, Sep 9, 2014 at 8:53 AM, Paul Grun <grun at cray.com> wrote:
> As a refresher, attached are the detailed minutes fromTuesday 8/19,,
> containing the results of the MPI point-to-point Pseudo-code working group.
> At our last meeting we had begun to work our way through the issues
> identified by that group.
> Regards,
> -Paul
> Cray Inc.
> Office:    (503) 620-8757
> Mobile:  (503) 703-5382
> _______________________________________________
> ofiwg mailing list
> ofiwg at lists.openfabrics.org
> http://lists.openfabrics.org/mailman/listinfo/ofiwg

Jeff Hammond
jeff.science at gmail.com

More information about the ofiwg mailing list