Compiling High Performance Fortran for Distributed-memory Systems

ming language for writing parallel programs. It is based on the Fortran 90 language, with extensions that enable the programmer to specify how array operations can be divided among multiple processors for increased performance. In HPF, the program specifies only the pattern in which the data is divided among the processors; the compiler automates the low-level details of synchronization and communication of data between processors. Digital’s DEC Fortran 90 compiler is the first implementation of the full HPF version 1.1 language (except for transcriptive argument passing, dynamic remapping, and nested FORALL and WHERE constructs). The compiler was designed for a distributedmemory machine made up of a cluster (or farm) of workstations and/or servers powered by Digital’s Alpha microprocessors. In a distributed-memory machine, communication between processors must be kept to an absolute minimum, because communication across the network is enormously more time-consuming than any operation done locally. Digital’s DEC Fortran 90 compiler includes a number of optimizations to minimize the cost of communication between processors. This paper briefly reviews the features of Fortran 90 and HPF that support parallelism, describes how the compiler implements these features efficiently, and concludes with some recent performance results showing that HPF programs compiled with Digital’s compiler yield performance that scales linearly or even superlinearly on significant applications on both distributed-memory and shared-memory architectures.