Evaluation of compiler optimizations for Fortran D on MIMD distributed memory machines
Seema Hiranandani, Ken Kennedy, Chau‐Wen Tseng · 1992
The Fortran D compiler uses data decomposition specifications to automatically translate Fortran programs for execution on MIMD distributed-memory machines. This paper introduces and classifies a number of advanced optimizations needed to achieve acceptable performance; they are analyzed and empirically evaluated for stencil computations. Profitability formulas are derived for each optimization. Results show that exploiting parallelism for pipelined computations, reductions, and scans is vital. Message vectorization, collective communication, and efficient coarsegrain pipelining also significantly affect performance. 1 Introduction Parallel computing represents the only plausible way to continue to increase the computational power available to computational scientists and engineers. However, parallel computers are not likely to be widely successful until they are also easy to program. MIMD distributed-memory machines such as the Intel iPSC/860 present the most difficult programming m...