This book is a well-recognized and leading guidebook on High Performance Computing for a broad audience of readers from industry and academia. The new edition is completely reorganized and updated to include the latest research in HPC. A new introductory chapter and a new chapter on accelerated computing have been added to the book to keep pace with recent advances.
Reihe
Auflage
Sprache
Verlagsort
Verlagsgruppe
Zielgruppe
Für Beruf und Forschung
Academic and Professional Practice & Development
Illustrationen
150 s/w Abbildungen
150 Illustrations, black and white
Maße
Höhe: 254 mm
Breite: 178 mm
ISBN-13
978-1-138-30445-1 (9781138304451)
Copyright in bibliographic data is held by Nielsen Book Services Limited or its licensors: all rights reserved.
Schweitzer Klassifikation
Autor*in
University of Erlangen Nuremberg, Germany
University of Erlangen Nuremberg, Germany
Introduction. Modern general-purpose processors. Basic optimization techniques for serial code. Parallel computers. Basics of parallelization. Shared-memory parallel programming with OpenMP. Node-level performance engineering. Efficient OpenMP programming. Locality optimizations on ccNUMA architectures. Distributed-memory parallel programming with MPI. Efficient MPI programming. Hybrid parallelization with MPI and OpenMP. Accelerated computing. Appendix: The LIKWID tool suite.