Publications of Torsten Hoefler

Oliver Fuhrer, T. Chadha, Torsten Hoefler, Grzegorz Kwasniewski, X. Lapillonne, D. Leutwyler, D. Luethi, Carlos Osuna, C. Schaer, Thomas Schulthess, Hannes Vogt:

Near-global climate simulation at 1 km resolution: establishing a performance baseline on 4888 GPUs with COSMO 5.0

(Geoscientific Model Development. Vol 11, Nr. 4, Copernicus Publications, May 2018)

Publisher Reference

Abstract

The best hope for reducing long-standing global climate model biases is by increasing resolution to the kilometer scale. Here we present results from an ultrahigh-resolution non-hydrostatic climate model for a near-global setup running on the full Piz Daint supercomputer on 4888 GPUs (graphics processing units). The dynamical core of the model has been completely rewritten using a domain-specific language (DSL) for performance portability across different hardware architectures. Physical parameterizations and diagnostics have been ported using compiler directives. To our knowledge this represents the first complete atmospheric model being run entirely on accelerators on this scale. At a grid spacing of 930 m (1.9 km), we achieve a simulation throughput of 0.043 (0.23) simulated years per day and an energy consumption of 596 MWh per simulated year. Furthermore, we propose a new memory usage efficiency (MUE) metric that considers how efficiently the memory bandwidth – the dominant bottleneck of climate codes – is being used.

Documents

download article:

BibTeX

@article{,
  author={Oliver Fuhrer and T. Chadha and Torsten Hoefler and Grzegorz Kwasniewski and X. Lapillonne and D. Leutwyler and D. Luethi and Carlos Osuna and C. Schaer and Thomas Schulthess and Hannes Vogt},
  title={{Near-global climate simulation at 1 km resolution: establishing a performance baseline on 4888 GPUs with COSMO 5.0}},
  journal={Geoscientific Model Development},
  year={2018},
  month={May},
  volume={11},
  number={4},
  publisher={Copernicus Publications},
  source={http://www.unixer.de/~htor/publications/},
}


serving: 18.191.181.231:62850	© Torsten Hoefler