Overview of the archetype implementation

To support this archetype, we provide a library of high-level routines that encapsulate the details of parallelism. This library facilitates more rapid development of applications. The current implementation is based on Fortran 77 (including a few of the most widely-supported extensions) and uses Fortran M [4,1] for parallelism. (An MPI-based implementation is also available.) To use the archetype implementation, the application developer must provide:

• A main program, typically consisting primarily of a call to spawn.
• Parameter definitions, as described in § 2.1 .
• A subroutine procmain that will be executed by each process. This subroutine will typically consist of sequential code much like that for a sequential version of the application, interspersed with calls to the archetype library routines described in § 3

  to perform those operations specific to the parallel version.

• User-specified and archetype-supplied parameters
• Grids and distributions
• Local data layout