All,

At the last meeting I attended (Dublin) we discussed developing a prototype "morph" ABI layer with initial implementation targeting the December meeting.  Let's take stock of this effort when we meet next week.  As I recall, Ezra offered to lead the windows implementation, and Alexander linux.  This exercise is meant to expose some of the implementation details and help drive us toward a detailed standard proposal we can take to the larger group.

Ezra and Alexander, please brief us on this effort when we meet next Wednesday.

From the Dublin meeting:
�MPI ABI standard, separate from MPI API standard, but under the umbrella of the MPI Forum
�Dynamic link is the main objective (binary compatibility)
�Startup (mpirun/mpiexec) is a secondary objective
�Include all major language bindings in the ABI standard (C, C++, Fortran) as we go
�Tied to MPI 2.1 and forward
�Implementers may choose to comply (or not)
�May be implemented as a morph layer or native
�Initial reference implementation will be a morph layer (tbd who provides)
�May be different per platform (Linux, Windows, �)
�Different voting rules separate from MPI API forum (?)
�Need participation from major implementors in order to make progress

mpi.h: MPICH 2 (1.0.7) as base with mods as needed (e.g. max values for compile time values used in declarations)
Proceed with a reference morph layer implementation targeting (at least) MPICH 2.0, OpenMPI
Load a different module and it just works
Demonstrate with a ring program
Defer startup consideration

Proof of concept reference morph layer implementations:
�Flesh out details by writing some code
�Linux on X86-64 cluster Red Hat Linux/SUSE
�Windows on X86-64 cluster
�Test against ring program
�Bring back to the December forum meeting

see ya next week

Jeff