Volume 6, Issue 3
Fall 1998

Volume 6, Issue 2
Spring/Summer 1998

Volume 6, Issue 1
Winter 1998

Volume 5, Issue 4
Fall 1997

Volume 5, Issue 3
Summer 1997

Volume 5, Issue 2
Spring 1997

Volume 5, Issue 1
Winter 1997

Volume 4, Issue 4
Fall 1996

Volume 4, Issue 3
Summer 1996

Volume 4, Issue 2
Spring 1996

Volume 4, Issue 1
Winter 1996

Volume 3, Issue 4
Fall 1995

Volume 3, Issue 3
Summer 1995

Volume 3, Issue 2
Spring 1995

Volume 3, Issue 1
January 1995

Volume 2, Issue 4
October 1994

Volume 2, Issue 3
July 1994

Volume 2, Issue 2
April 1994

Volume 2, Issue 1
January 1994

Volume 1, Issue 4
October 1993

Volume 1, Issue 3
July 1993

Volume 1, Issue 2
April 1993

Volume 1, Issue 1
January 1993

Since its introduction over three decades ago, Fortran has been the language of choice for high-level scientific programming of sequential computers. It has strong support in the scientific community and a huge body of existing applications. However, ordinary Fortran 77 and Fortran 90 do not exploit the full capacity of modern computer architectures, which increasingly require more program information in areas such as:

• Opportunities for parallel execution
• Type of available parallelism - MIMD, SIMD, or some combination
• Allocation of data among individual processor memories

To address these problems, a coalition of academic and industrial groups has formed the High Performance Fortran Forum (HPFF) under the direction of Ken Kennedy of the CRPC. Chuck Koelbel, also of the CRPC, is serving as executive director. The HPFF's goal is to develop a set of standard extensions to Fortran that provide the necessary information to support high performance programming on a wide variety of platforms, including massively parallel SIMD and MIMD systems, vector processors, and RISC processors with complex memory hierarchies. Based on language proposals from COMPASS and DEC (High Performance Fortran), Cray Research (MPP Programming Model), Rice and Syracuse Universities (Fortran D), and the University of Vienna (Vienna Fortran), the new language, called "High Performance Fortran," will include:

• Constructs for array-based parallelism, such as the Fortran 90 vector operations
• Constructs for specification of parallel loops and complex parallel assignments
• Constructs for specifying complex parallel assignments and other operations on large data
• Constructs for specifying the allocation and placement of data for efficient access

The result of this effort will be an industry-wide (informal) standard for a language that is portable from workstations to massively parallel supercomputers. Furthermore, the language will be able to express the algorithms needed to achieve high performance on specific architectures in this range.

The HPF language specification version 1.0 is now available. The High Performance Fortran Forum met every six weeks beginning in January 1992 with the goal of producing a draft by December 1992. Version 1.0 of the language specification is now available, and HPFF is seeking public comments on its technical merits. (See the "Request for Public Comment" article.)

Geoffrey Fox of the CRPC is leading an effort to develop a test suite for High Performance Fortran. He has also been successful in collaborating with corporate researchers to develop commercial spin-offs of HPF. On November 16, 1992, Fox and researchers from the Portland Group, Inc., announced the formation of a technology alliance to research, develop, and distribute commercial parallel compilers and tools based on the HPF language (See related article).