Accelerating tracer transport in FESOM-2 with GPU's

Within the ESiWACE-2 service on porting applications to pre-exascale machines, we have collaborated with the FESOM-2 team to accelerate parts of their model on GPU's. FESOM-2 is a finite-volume ocean and sea-ice model based on unstructured meshes. We have used the OpenACC programming model to s...

Full description

Bibliographic Details
Main Authors: Gijs van den Oord, David Guibert, Alessio Sclocco, Erwan Raffin, Ben van Werkhoven, Natalya Rakowski, Nicolay Koldunov, Dmitry Sidorenko
Format: Conference Object
Language:English
Published: 2022
Subjects:
Ocean modeling
High-performance computing
GPU
Sea ice
Online Access:https://zenodo.org/record/6580762
https://doi.org/10.5281/zenodo.6580762
Description
Summary:Within the ESiWACE-2 service on porting applications to pre-exascale machines, we have collaborated with the FESOM-2 team to accelerate parts of their model on GPU's. FESOM-2 is a finite-volume ocean and sea-ice model based on unstructured meshes. We have used the OpenACC programming model to speed up the transport of tracers in FESOM-2. In this talk, I will share the results of this effort, describe our collaboration with the FESOM team and present some technical aspects of the implementation. Finally, I will iterate on the lessons learned and the road ahead for FESOM-2 towards exascale.

Similar Items