Redirigiendo al acceso original de articulo en 15 segundos...
ARTÍCULO
TITULO

Performance Portability of HPC Discovery Science Software: Fusion Energy Turbulence Simulations at Extreme Scale

William Tang    
Bei Wang    
Stephane Ethier    
Zhihong Lin    

Resumen

As HPC R&D moves forward on a variety of ?path to exascale? architectures today, an associated objective is to demonstrate performance portability of discovery-science-capable software.  Important application domains, such as Magnetic Fusion Energy (MFE), have improved modelling of increasingly complex physical systems -- especially with respect to reducing ?time-to-solution? as well as  ?energy to solution.?  The emergence of new insights on confinement scaling in MFE systems has been aided significantly by efficient software capable of harnessing powerful supercomputers to carry out simulations with unprecedented resolution and temporal duration to address increasing problem sizes.  Specifically, highly scalable particle-in-cell (PIC) programing methodology is used in this paper to demonstrate how modern scientific applications can achieve efficient architecture-dependent optimizations of performance scaling and code portability for path-to-exascale platforms.

 Artículos similares

       
 
Fatemeh Roshan, Sasan Tavakoli, Simone Mancini and Abbas Dashtimanesh    
A tunneled planing craft is a high-speed boat with two tunnels over the hull bottom that are designed to improve the vessel?s performance. Hydrodynamic performance of tunneled planing hulls in calm-water is well-known, however, current information on wav... ver más

 
Wei Wang, Xuefeng Hong, Jue Qu, Ning Xu and Tong Chen    
A complex command and control task was selected as the test task, which included observing the overall and local situation, the interactive operation and situation display of detection equipment, the erection and launch of air defense equipment, and the ... ver más
Revista: Aerospace

 
Aurelien Bloch, Simone Casale-Brunet and Marco Mattavelli    
Developing and fine-tuning software programs for heterogeneous hardware such as CPU/GPU processing platforms comprise a highly complex endeavor that demands considerable time and effort of software engineers and requires evaluating various fundamental co... ver más

 
Mirco De Marchi, Francesco Lumpp, Enrico Martini, Michele Boldo, Stefano Aldegheri and Nicola Bombieri    
Many modern programmable embedded devices contain CPUs and a GPU that share the same system memory on a single die. Such a unified memory architecture (UMA) allows programmers to implement different communication models between CPU and the integrated GPU... ver más

 
Hao-Yi Yang, Zhi-Rong Lin and Ko-Chih Wang    
The use of distribution-based data representation to handle large-scale scientific datasets is a promising approach. Distribution-based approaches often transform a scientific dataset into many distributions, each of which is calculated from a small numb... ver más
Revista: Algorithms