Program Technical Status

A. Description of progress during the reporting period, supported by
reasons for any changes reported previously.

Work continues according to plan.  Highlights:

  The first public domain release of the PCRC kernel software,
  (developed at NPAC) is now expected in the next reporting period.
  The functionality of the kernel is essentially frozen and fully
  implemented.  The interface and implementation are currently being
  documented, prior to release.  The final task before release will be
  to simplifiy the installation procedures, improving portability
  across platforms with diverse operating systems, MPI implementations,
  and compilers.  A GNU-style configuration script is under
  development.

  The proposed Java interface to the kernel runtime is being actively
  developed.  We have devised a minimal set of extensions to the Java
  language to support data-parallel programming using distributed arrays
  plus the PCRC run-time functions.  Besides its intrinsic value
  as a Java environment for parallel programming, the envisaged system
  can be viewed as a pilot study towards incorporation of similar
  extensions in more complex languages.  This work comes under the
  umbrella of stage 2, item 4---{\em Infrastructure supporting high-level
  SPMD}, and stage 1, item 1g---{\em Java interface} in the
  revised statement of work, submitted with the previous quarterly
  report.

  The Florida subcontractors have developed an MPI version of their
  algorithms for combining scatter.  This version is link-compatible with
  the PCRC kernel developed at Syracuse.  Syracuse has provided interface
  code so that these routines are now callable using the generic DAD
  array descriptors employed by the kernel library.

  The work of subcontractors Texas and Maryland, developing DAGH and
  MetaChaos respectively, is proceeding according to plan.

  Something on prefix.

  Excuse for a compiler.

  ... Should address things in revised statement of work for this period.

B. Description of any special programs developed or major items of
experimental or special equipment purchased or constructed during the
reporting period.

None.

C. Notification of any changes in key personnel associated with the
   contract during the reporting period.

No change.

D. Summary of substantive information derived from noteworthy trips,
   meetings, and special conferences held in connection with the contract
   during the reporting period.

   1) Papers ``PCRC-based HPF Compilation'' and ``Java as a Language
      for Scientific Parallel Programming'' were presented at LCPC '97
      in Minnesota.

   ...

E. Summary of any problems or areas of concern on which procuring
   activity or guidance is required.

None.

F. Statement relative to apparent sufficiency of contractually prescribed
   effort to achieve the objectives of the contract.

The contractually prescribed effort is sufficient to achieve the
objectives of this contract.

G. Related accomplishments: i.e., scientific papers published, etc.

   1) PCRC-based HPF Compilation

   2) Java as a Language for Scientific Parallel Programming

   ... HPJava proposals???

   ... Seungjo+Sanjay report

H. Planned activities and milestones for the next reporting period.

   The documentation of the NPAC runtime kernel will be completed.
   (This is a substantial document.  The current draft---some 120
   pages---covers about half of the system.)

   The runtime kernel will be tested on various UNIX platforms,
   (including LINUX).  The installation scripts will be completed.

   The release of the kernel software will be made public.

   Java interface.  A Java interface complementing the current
   C++ and Fortran interfaces will be completed.  Work will start
   on a source-to-source preprocessor for a minimally extended Java dialect,
   providing a high-level interface to the PCRC run-time (this can be
   seen as a new compiler targetting the run-time---an {\em
   HPJava}---although the scope of this work is somewhat less ambitious
   than for HPF).  As explained in section A, this work fits naturally
   in the revised work plan.  It will exploit an existing
   infra-structure for Java compilation.

   ... Any things from the revised SOW that fit naturally into this time frame.