Results (1)
Search Parameters:
Keyword: Performance portabilityOpen AccessArticle
9 Pages, 631 KB 
Performance Portability and Unified Profiling for Finite Element Methods on Parallel Systems
Abstract:
The currently available variety of modern, highly-parallel universal processors includes multi-core CPU and many-core GPU (Graphics Processing Units) from different vendors. Systems composed of such processors enable high-performance execution of demanding applications like numerical Finite Element Methods. However, today’s application pro- gramming for parallel systems lacks performance portability: the same program code cannot achieve stable…
Read More(This article belongs to the SP8 (Special Issue on Multidisciplinary Sciences and Engineering 2019-20) & Section Information Systems in Computer Science (CIS))