Experience in the Automatic Parallelization of Four Perfect-Benchmark Programs
نویسندگان
چکیده
This paper discusses the techniques used to hand-parallelize, for the Alliant FX/80, four Fortran programs from the Perfect-Benchmark suite. The paper also includes the execution times of the programs before and after the transformations. The four programs considered here were not effectively parallelized by the automatic translators available to the authors. However, most of the techniques used for hand parallelization, and perhaps all of them, have wide applicability and can be incorporated into existing translators.
منابع مشابه
Implementation and evaluation of HPF/SX V2
We are developing HPF/SX V2, an HPF compiler for vector parallel machines. It provides some unique extensions as well as the features of HPF 2.0 and HPF/JA. This paper describes in particular four of them: 1) the ON directive of HPF 2.0, 2) the REFLECT and LOCAL directives of HPF/JA, 3) vectorization directives, and 4) automatic parallelization. We evaluated these features through some benchmar...
متن کاملAutomatic Parallelization Using the Value Evolution Graph
We introduce a framework for the analysis of memory reference sets addressed by induction variables without closed forms. This framework relies on a new data structure, the Value Evolution Graph (VEG), which models the global flow of scalar and array values within a program. We describe the application of our framework to array dataflow analysis, privatization, and dependence analysis. This res...
متن کاملAutomatic Loop Parallelization: An Abstract Interpretation Approach
This paper presents an abstract interpretation to support the automatic parallelization of iterative constructs. The analysis approximates the range of values of any array subscript by integrating an interval analysis with one returning a set of simple linear restraints among program variables. To reduce the complexity of the analysis, a general methodology has been adopted that defines two dif...
متن کاملInterprocedural Parallelization Analysis: Preliminary Results
This paper describes a fully interprocedural automatic parallelization system for Fortran programs, and presents the results of extensive experiments obtained using this system. The system incorporates a comprehensive and integrated collection of analyses including dependence, privatization and reduction recognition for both array and scalar variables, and scalar symbolic analysis to support th...
متن کاملPerformance Analysis of Parallelizing Compilers on the Perfect BenchmarksTM Programs
We have studied the eeectiveness of parallelizing compilers and the underlying transformation techniques. This paper reports the speedups of the Perfect Benchmarks TM codes that result from automatic parallelization. We have further measured the performance gains caused by individual restructuring techniques. Speciic reasons for the successes and failures of the transformations are discussed, a...
متن کامل