A Portable Multidimensional Coarray for C++
Authors:
- Felix MoBbauerFMFelix MoBbauer
- Roger KowalewskiRKRoger Kowalewski
- Tobias FuchsTFTobias Fuchs
- Karl FurlingerKFKarl Furlinger
- 1 March 2018
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 18-25
Abstract
Fortran Coarrays are a well known data structure in High Performance Computing (HPC) applications. There have been various attempts to port the concept to other programming languages that have a wider user base outside of scientific computing. While a popular implementation of the partitioned global address space (PGAS) model is Unified Parallel C (UPC), there is currently no portable implementation of Coarrays for C++. In this paper a portable version is presented, which is closely based on the Coarray C++ implementation of the Cray Compiling Environment. In this work we focus on a common subset of all proposed features by Cray. Our implementation utilizes the distributed data structures provided by the DASH library, demonstrating their universal applicability. Finally, a performance evaluation shows that our proposed Coarray abstraction adds negligible overhead and even outperforms native Coarray Fortran.Keywords
This publication has 15 references indexed in Scilit:
- A Multi-dimensional Distributed Array Abstraction for PGAS2016
- Expressing and Exploiting Multi-Dimensional Locality in DASH2016
- CoGA: Extension of GA on Heterogeneous System2015
- DART-MPI2014
- UPC++: A PGAS Extension for C++2014
- GA-GPU2012
- Parallel Programmability and the Chapel LanguageThe International Journal of High Performance Computing Applications, 2007
- X10ACM SIGPLAN Notices, 2005
- ARMCI: A portable remote memory copy library for distributed array libraries and compiler run-time systems1999
- Global arrays: A nonuniform memory access programming model for high-performance computersThe Journal of Supercomputing, 1996