Mitsuhisa Sato

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.

From Programming Model to Applications

XcalableMP PGAS Programming Language

Kazuki Joe

Hans P. Zima

Yoshiki Seo

Masaaki Shimasaki

I wish to welcome all of you to the International Symposium on High Perf- mance Computing 2002 (ISHPC2002) and to Kansai Science City, which is not farfromtheancientcapitalsofJapan:NaraandKyoto.ISHPC2002isthefourth in the ISHPC series, which consists, to date, of ISHPC ’97 (Fukuoka, November 1997), ISHPC ’99 (Kyoto, May 1999), and ISHPC2000 (Tokyo, October 2000). The success of these symposia indicates the importance of this area and the strong interest of the research community. With all of the recent drastic changes in HPC technology trends, HPC has had and will continue to have a signi?cant impact on computer science and technology. I am pleased to serve as General Chair at a time when HPC plays a crucial role in the era of the IT (Information Technology) revolution. The objective of this symposium is to exchange the latest research results in software, architecture, and applications in HPC in a more informal and friendly atmosphere. I am delighted that the symposium is, like past successful ISHPCs, comprised of excellent invited talks, panels, workshops, as well as high-quality technical papers on various aspects of HPC. We hope that the symposium will provide an excellent opportunity for lively exchange and discussion about - rections in HPC technologies and all the participants will enjoy not only the symposium but also their stay in Kansai Science City.

4th International Symposium, ISHPC 2002, Kansai Science City, Japan, May 15-17, 2002. Proceedings

High Performance Computing

Covering research topics from system software such as programming languages, compilers, runtime systems, operating systems, communication middleware, and large-scale file systems, as well as application development support software and big-data processing software, this book presents cutting-edge software technologies for extreme scale computing. The findings presented here will provide researchers in these fields with important insights for the further development of exascale computing technologies. This book grew out of the post-peta CREST research project funded by the Japan Science and Technology Agency, the goal of which was to establish software technologies for exploring extreme performance computing beyond petascale computing. The respective were contributed by 14 research teams involved in the project. In addition to advanced technologies for large-scale numerical computation, the project addressed the technologies required for big data and graph processing, the complexity of memory hierarchy, and the power problem. Mapping the direction of future high-performance computing was also a central priority.

Mitsuhisa Sato

About

Books

Join the club