In EDS ansehen

Research on key techniques for parallelization and optimization of applied codes

Gespeichert in:
Bibliographische Detailangaben
Titel: Research on key techniques for parallelization and optimization of applied codes
Autoren: Mo, Zeyao, Zhang, Aiqing, Zhang, Xia
Verlagsinformationen: Allerton Press, New York, NY
Schlagwörter: Other programming paradigms (object-oriented, sequential, concurrent, automatic, etc.), MPI, OpenMP, parallel programming paradigms
Beschreibung: Summary: Based on the high performance characteristics of the hierarchical memory existed in the state-of-the-art powerful microprocessors and the two leading parallel programming paradigms including MPI and OpenMP supported widely by parallel computers, this paper summarized and discussed some key techniques for parallelization and optimization of seven typical applied codes with different physical model problems. Typical benchmark results under six parallel computers are also given in detail in this paper.
Publikationsart: Article
Dateibeschreibung: application/xml
Zugangs-URL: https://zbmath.org/1950371
Dokumentencode: edsair.c2b0b933574d..ab2b98db77d898e44c1bda99943f5a3f
Datenbank: OpenAIRE
Beschreibung
Abstract:Summary: Based on the high performance characteristics of the hierarchical memory existed in the state-of-the-art powerful microprocessors and the two leading parallel programming paradigms including MPI and OpenMP supported widely by parallel computers, this paper summarized and discussed some key techniques for parallelization and optimization of seven typical applied codes with different physical model problems. Typical benchmark results under six parallel computers are also given in detail in this paper.