A distributed-memory tridiagonal solver based on a specialised data structure optimised for CPU and GPU architectures

Various numerical methods used for solving partial differential equations (PDE) result in tridiagonal systems. Solving tridiagonal systems on distributed-memory environments is not straightforward, and often requires significant amount of communication. In this article, we present a novel distribute...

Celý popis

Podrobná bibliografie

Vydáno v:	Computer physics communications Ročník 315; s. 109747
Hlavní autoři:	Akkurt, Semih, Lemaire, Sébastien, Bartholomew, Paul, Laizet, Sylvain
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	Elsevier B.V 01.10.2025
Témata:	Compact finite difference schemes Distributed tridiagonal matrix solver Tridiagonal matrix algorithm Distributed tridiagonal matrix solver Compact finite difference schemes Tridiagonal matrix algorithm
ISSN:	0010-4655
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!