Endre søk
RefereraExporteraLink to record
Permanent link

Direct link
Referera
Referensformat
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annet format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annet språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf
Enabling mixed-precision in spectral element codes
Umeå University, Sweden.
Université Paris-Saclay, UVSQ, LI-PaRAD, France.
Umeå University, Sweden.
KTH, Skolan för elektroteknik och datavetenskap (EECS), Centra, Parallelldatorcentrum, PDC.ORCID-id: 0000-0002-5020-1631
Vise andre og tillknytning
2026 (engelsk)Inngår i: Future Generation Computer Systems, ISSN 0167-739X, E-ISSN 1872-7115, Vol. 174, artikkel-id 107990Artikkel i tidsskrift (Fagfellevurdert) Published
Abstract [en]

Mixed-precision computing has the potential to significantly reduce the cost of exascale computations, but determining when and how to implement it in programs can be challenging. In this article, we propose a methodology for enabling mixed-precision with the help of computer arithmetic tools, roofline model, and computer arithmetic techniques. As case studies, we consider Nekbone (Nek5000 developers), a mini-application for the Computational Fluid Dynamics (CFD) solver Nek5000 (Fischer et al.), and a modern Neko (Jansson et al., 2024) CFD application. With the help of the Verificarlo (Denis et al., 2016) tool and computer arithmetic techniques, we introduce a strategy to address stagnation issues in the preconditioned Conjugate Gradient method in Nekbone and apply these insights to implement a mixed-precision version of Neko. We evaluate the derived mixed-precision versions of these codes by combining metrics in three dimensions: accuracy, time-to-solution, and energy-to-solution. Notably, mixed-precision in Nekbone reduces time-to-solution by roughly 1.62x and energy-to-solution by 2.43x on MareNostrum 5, while in the real-world Neko application, the gain is up to 1.3x in both time and energy, with the accuracy that matches double-precision results.

sted, utgiver, år, opplag, sider
Elsevier BV , 2026. Vol. 174, artikkel-id 107990
Emneord [en]
Computer arithmetic tool, Conjugate gradient, Energy-to-solution, Mixed-precision, Neko, Roofline model, Verificarlo
HSV kategori
Identifikatorer
URN: urn:nbn:se:kth:diva-368935DOI: 10.1016/j.future.2025.107990ISI: 001528005900003Scopus ID: 2-s2.0-105009726439OAI: oai:DiVA.org:kth-368935DiVA, id: diva2:1992894
Merknad

QC 20250828

Tilgjengelig fra: 2025-08-28 Laget: 2025-08-28 Sist oppdatert: 2025-11-13bibliografisk kontrollert

Open Access i DiVA

Fulltekst mangler i DiVA

Andre lenker

Forlagets fulltekstScopus

Person

Jansson, Niclas

Søk i DiVA

Av forfatter/redaktør
Jansson, Niclas
Av organisasjonen
I samme tidsskrift
Future Generation Computer Systems

Søk utenfor DiVA

GoogleGoogle Scholar

doi
urn-nbn

Altmetric

doi
urn-nbn
Totalt: 141 treff
RefereraExporteraLink to record
Permanent link

Direct link
Referera
Referensformat
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annet format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annet språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf