New Mathematics for Computer Performance: Array Algebra and Cost Functions
Mathematics · 28.04.2026 · 14 (9) · 1479
DOI: 10.3390/math14091479 ↗Kurz gefasst
Fachartikel in der Zeitschrift Mathematics, online erschienen am 28.04.2026. Verfasst von Gaétan Hains und 1 weiteren Autor:innen. Die Arbeit nennt 42 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NSF USA, CRIANN.
Abstract (Originalsprache)
MoA (mathematics of arrays) is a theory of parallel operations on arrays that can describe all known algorithms in linear algebra, signal processing, and HPC because they are based on primitive recursion and array shapes. Mapping parallel algorithms to computer architectures remains more of an art than a science, and specific mathematical techniques are needed to provide a basis for performance evaluation at a level abstract enough to constitute an experimental science. In this paper we present a methodology for parallel code generation from MoA expressions. Then, we relate the MoA operators to the linear space of memory elements in computer architecture. Finally, we define a theory of execution costs that is based on classical operations research and is formally related to MoA-based parallel code generation. This constitutes a formalized and mechanizable approach to performance prediction, portability and optimization.
Förderung
- NSF USA
- CRIANN
Zitieren
APA
Hains, G., & Mullin, L. (2026). New Mathematics for Computer Performance: Array Algebra and Cost Functions. Mathematics, 14(9), 1479. https://doi.org/10.3390/math14091479
Vancouver
Hains G, Mullin L. New Mathematics for Computer Performance: Array Algebra and Cost Functions. Mathematics. 2026;14(9):1479. doi:10.3390/math14091479
BibTeX
@article{Hains2026,
author = {Hains, Gaétan and Mullin, Lenore},
title = {New Mathematics for Computer Performance: Array Algebra and Cost Functions},
journal = {Mathematics},
year = {2026},
volume = {14},
number = {9},
pages = {1479},
doi = {10.3390/math14091479}
}
EUVIMED