Automated generation of high-performance computational fluid dynamics codes
Visualitza/Obre
Cita com:
hdl:2117/367725
Tipus de documentArticle
Data publicació2022-05
EditorElsevier
Condicions d'accésAccés obert
Llevat que s'hi indiqui el contrari, els
continguts d'aquesta obra estan subjectes a la llicència de Creative Commons
:
Reconeixement-NoComercial-SenseObraDerivada 4.0 Internacional
ProjecteDEEP-SEA - DEEP – SOFTWARE FOR EXASCALE ARCHITECTURES (EC-H2020-955606)
UPC-COMPUTACION DE ALTAS PRESTACIONES VIII (AEI-PID2019-107255GB-C22)
BARCELONA SUPERCOMPUTING CENTER - CENTRO. NACIONAL DE SUPERCOMPUTACION (MINECO-SEV-2015-0493)
UPC-COMPUTACION DE ALTAS PRESTACIONES VIII (AEI-PID2019-107255GB-C22)
BARCELONA SUPERCOMPUTING CENTER - CENTRO. NACIONAL DE SUPERCOMPUTACION (MINECO-SEV-2015-0493)
Abstract
Domain-Specific Languages (DSLs) improve programmers productivity by decoupling problem descriptions from algorithmic implementations. However, DSLs for High-Performance Computing (HPC) have two additional critical requirements: performance and scalability. This paper presents the automated process of generating, from abstract mathematical specifications of Computational Fluid Dynamics (CFD) problems, optimised parallel codes that perform and scale as manually optimised ones. We consciously combine within Saiph, a DSL for solving CFD problems, low-level optimisations and parallelisation strategies, enabling high-performance single-core executions which effectively scale to multi-core and distributed environments. Our results demonstrate how high-level DSLs can offer competitive performance by transparently leveraging state-of-the-art HPC techniques.
CitacióMacià, S. [et al.]. Automated generation of high-performance computational fluid dynamics codes. "Journal of computational science", Maig 2022, vol. 61, article 101664.
ISSN1877-7503
Versió de l'editorhttps://www.sciencedirect.com/science/article/pii/S1877750322000758
Altres identificadorshttps://arxiv.org/abs/2204.12120
Fitxers | Descripció | Mida | Format | Visualitza |
---|---|---|---|---|
Macia et al.pdf | 1,695Mb | Visualitza/Obre |