An improved automatic MPI code generation algorithm for parallelizing compilation

Open64 is an open source compiler with powerful analysis and widely used as a research and commercial development platform. However, it has not been designed and developed to realize MPI parallelization. There are many contributions in the paper. Firstly, the Open64 compiler infrastructure is showed...

Full description

Saved in:
Bibliographic Details
Published in2017 IEEE 2nd Advanced Information Technology, Electronic and Automation Control Conference (IAEAC) pp. 1623 - 1626
Main Authors Yangxia Xiang, Caisen Chen, Hongyan Wang, Zeyun Zhou
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.03.2017
Subjects
Arrays
Benchmark testing
Clustering algorithms
code generation
Generators
Kernel
MPI
Open64 compiler
parallelizing compilation
Program processors
Online AccessGet full text
DOI10.1109/IAEAC.2017.8054288

Cover

More Information
Summary:Open64 is an open source compiler with powerful analysis and widely used as a research and commercial development platform. However, it has not been designed and developed to realize MPI parallelization. There are many contributions in the paper. Firstly, the Open64 compiler infrastructure is showed. Secondly, the location of MPI code generation in the Open64 compiler architecture is analyzed. Thirdly, an Open64-based automatic generation algorithm for MPI code is presented. By the experiments of testing the NPB benchmarks, the results show that the approach not only has a high rate of parallel transformation, but also can produce correct MPI parallel programs.
DOI:10.1109/IAEAC.2017.8054288