COMPILING FOR MASSIVELY-PARALLEL ARCHITECTURES - A PERSPECTIVE

The problem of automatically generating programs for massively paralle l computers is a very complicated one, mainly because there are many a rchitectures, each of them seeming to pose its own particular compilat ion problem. The purpose of this paper is to propose a framework in wh ich to discuss the compilation process, and to show that the features which affect it are few and generate a small number of combinations. T he paper is oriented toward fine-grained parallelization of static con trol programs, with emphasis on dataflow analysis, scheduling and plac ement. When going from there to more general programs and to coarser p arallelism, one encounters new problems, some of which are discussed i n the conclusion.