Implementation And Evaluation Of Additional Parallel Features In Coarray Fortran
Download Implementation And Evaluation Of Additional Parallel Features In Coarray Fortran PDF/ePub or read online books in Mobi eBooks. Click Download or Read Online button to get Implementation And Evaluation Of Additional Parallel Features In Coarray Fortran book now. This website allows unlimited access to, at the time of writing, more than 1.5 million titles, including hundreds of thousands of titles in various foreign languages.
Implementation and Evaluation of Additional Parallel Features in Coarray Fortran
The Fortran 2008 language standard added a feature called "coarrays" to allow parallel programming in Fortran with only minimal changes to existing sequential Fortran programs. Coarrays turn Fortran into a Partitioned Global Address Space (PGAS) language, following the Single Program, Multiple Data (SPMD) model. The next revision of the Fortran standard is expected to introduce some more sophisticated coarrays language features. One feature is the "team"; a way of grouping components (images) of parallel Fortran programs. Teams can, for example, be allocated different sub-tasks. Proposed team support in the standard includes statements for forming image teams, reassigning membership of teams, and statements for performing communication and synchronization with respect to image teams. These features are collected and discussed in the Fortran Technical Specification Draft. In this thesis, we will present implementation and evaluation of some of these new features. The open-source compiler, OpenUH, developed by this research group is extended to implement support for team and collective. We discuss two optimizations we have applied in order to reduce network communication and local memory footprint in the compiler's Coarrays runtime. Experimental results using several micro-benchmarks, one benchmark from the NAS Parallel Benchmark suite and High Performance Linpack suite show that new features make the program logic more concise, while achieving good performance.
XcalableMP PGAS Programming Language
XcalableMP is a directive-based parallel programming language based on Fortran and C, supporting a Partitioned Global Address Space (PGAS) model for distributed memory parallel systems. This open access book presents XcalableMP language from its programming model and basic concept to the experience and performance of applications described in XcalableMP. XcalableMP was taken as a parallel programming language project in the FLAGSHIP 2020 project, which was to develop the Japanese flagship supercomputer, Fugaku, for improving the productivity of parallel programing. XcalableMP is now available on Fugaku and its performance is enhanced by the Fugaku interconnect, Tofu-D. The global-view programming model of XcalableMP, inherited from High-Performance Fortran (HPF), provides an easy and useful solution to parallelize data-parallel programs with directives for distributed global array and work distribution and shadow communication. The local-view programming adopts coarray notation from Coarray Fortran (CAF) to describe explicit communication in a PGAS model. The language specification was designed and proposed by the XcalableMP Specification Working Group organized in the PC Consortium, Japan. The Omni XcalableMP compiler is a production-level reference implementation of XcalableMP compiler for C and Fortran 2008, developed by RIKEN CCS and the University of Tsukuba. The performance of the XcalableMP program was used in the Fugaku as well as the K computer. A performance study showed that XcalableMP enables a scalable performance comparable to the message passing interface (MPI) version with a clean and easy-to-understand programming style requiring little effort.
Coarray Programming in Fortran
"Coarray Programming in Fortran" "Coarray Programming in Fortran" is an authoritative and comprehensive exploration of coarray parallelism in one of the world's most enduring scientific programming languages. Beginning with a historical perspective, the book traces Fortran’s evolution in parallel computing and delves into the foundational principles of the Partitioned Global Address Space (PGAS) model, illustrating how coarrays extend Fortran's capabilities for high-performance, scalable computing. Readers are guided through the modern standards from Fortran 2008 to Fortran 2018, and presented with thoughtful comparisons between coarrays and other parallel models such as MPI and OpenMP, highlighting the unique strengths and application domains enabled by coarray-based design. The book offers a meticulous tour of coarray syntax, semantics, and advanced programming patterns. Practical guidance is given on all essential topics including image management, synchronization, robust communication algorithms, and the construction of sophisticated parallel data structures. Deep, hands-on insights cover both the technical details—such as dynamic allocation, remote memory access, and atomic operations—and practical engineering challenges, from performance optimization and debugging to leveraging hybrid programming models across heterogeneous systems. Real-world code examples and case studies weave theoretical knowledge into applied best practices, ensuring readers are well-equipped for both correctness and efficiency in parallel application development. Looking beyond established territory, the concluding chapters survey the ongoing evolution and exciting future of coarray programming. Coverage includes emerging standards, new application domains like exascale computing and AI, and the integration of coarrays in modern cloud and containerized environments. The book also emphasizes community-driven progress and enduring research challenges, making it an invaluable reference for scientists, engineers, and researchers committed to pushing the boundaries of scalable parallel computing with Fortran.