E3SM-MMF: Cloud Resolving Model Development on Heterogeneous Supercomputers

Cloud Resolving Model (CRM) development for efficient utilization of GPUs on hybrid supercomputers in preparation for exascale. This work is part of a larger project, E3SM-MMF component of the U.S. Department of Energy(DOE) Exascale Computing Project. The E3SM-MMF approach addresses structural uncer...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Norman, Matt, Sreepathi, Sarat, Anikesh Pal, Hannah, Walter, Taylor, Mark
Format: Bild
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Cloud Resolving Model (CRM) development for efficient utilization of GPUs on hybrid supercomputers in preparation for exascale. This work is part of a larger project, E3SM-MMF component of the U.S. Department of Energy(DOE) Exascale Computing Project. The E3SM-MMF approach addresses structural uncertainty in cloud processes by replacing traditional parameterizations with cloud resolving “super-parameterization” within each grid cell of global climate model. Super-parameterization dramatically increases arithmetic intensity, making the MMF approach an ideal strategy to achieve good performance on emerging exascale computing architectures.
DOI:10.6084/m9.figshare.6262097