Articles | Volume 12, issue 7
https://doi.org/10.5194/gmd-12-3135-2019
© Author(s) 2019. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
https://doi.org/10.5194/gmd-12-3135-2019
© Author(s) 2019. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Development and technical paper
| 
24 Jul 2019
Development and technical paper |
| 24 Jul 2019
| 24 Jul 2019
How to use mixed precision in ocean models: exploring a potential reduction of numerical precision in NEMO 4.0 and ROMS 3.6
Oriol Tintó Prims
Earth Sciences Department, Barcelona Supercomputing Center – Centro Nacional de Supercomputación, Barcelona, Spain
HPCA4SE research group, Computer Architecture and Operating Systems Department, Universitat Autònoma de Barcelona, Bellaterra, Spain
Mario C. Acosta
Earth Sciences Department, Barcelona Supercomputing Center – Centro Nacional de Supercomputación, Barcelona, Spain
Andrew M. Moore
Ocean Sciences Department, University of California, Santa Cruz, CA, USA
Miguel Castrillo
Earth Sciences Department, Barcelona Supercomputing Center – Centro Nacional de Supercomputación, Barcelona, Spain
Kim Serradell
Earth Sciences Department, Barcelona Supercomputing Center – Centro Nacional de Supercomputación, Barcelona, Spain
Ana Cortés
HPCA4SE research group, Computer Architecture and Operating Systems Department, Universitat Autònoma de Barcelona, Bellaterra, Spain
Francisco J. Doblas-Reyes
Earth Sciences Department, Barcelona Supercomputing Center – Centro Nacional de Supercomputación, Barcelona, Spain
ICREA, Barcelona, Spain
Viewed
Total article views: 2,776 (including HTML, PDF, and XML)
Cumulative views and downloads
(calculated since 15 Feb 2019)
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 1,853 | 878 | 45 | 2,776 | 88 | 64 |
- HTML: 1,853
- PDF: 878
- XML: 45
- Total: 2,776
- BibTeX: 88
- EndNote: 64
Total article views: 2,098 (including HTML, PDF, and XML)
Cumulative views and downloads
(calculated since 24 Jul 2019)
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 1,465 | 590 | 43 | 2,098 | 80 | 56 |
- HTML: 1,465
- PDF: 590
- XML: 43
- Total: 2,098
- BibTeX: 80
- EndNote: 56
Total article views: 678 (including HTML, PDF, and XML)
Cumulative views and downloads
(calculated since 15 Feb 2019)
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 388 | 288 | 2 | 678 | 8 | 8 |
- HTML: 388
- PDF: 288
- XML: 2
- Total: 678
- BibTeX: 8
- EndNote: 8
Viewed (geographical distribution)
Total article views: 2,776 (including HTML, PDF, and XML)
Thereof 2,447 with geography defined
and 329 with unknown origin.
Total article views: 2,098 (including HTML, PDF, and XML)
Thereof 1,867 with geography defined
and 231 with unknown origin.
Total article views: 678 (including HTML, PDF, and XML)
Thereof 580 with geography defined
and 98 with unknown origin.
| Country | # | Views | % |
|---|
| Country | # | Views | % |
|---|
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
1
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
1
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
1
Cited
16 citations as recorded by crossref.
- Compressing atmospheric data into its real information content M. Klöwer et al. 10.1038/s43588-021-00156-2
- Reduced-Precision Chemical Kinetics in Atmospheric Models K. Sophocleous & T. Christoudias 10.3390/atmos13091418
- Heterogeneous acceleration algorithms for shallow cumulus convection scheme over GPU clusters F. Li et al. 10.1016/j.future.2023.04.021
- Discussion on “Saving Storage in Climate Ensembles: A Model-Based Stochastic Approach” J. Bessac et al. 10.1007/s13253-023-00540-7
- Numerical algorithms for high-performance computational science J. Dongarra et al. 10.1098/rsta.2019.0066
- An Objective and Efficient Method for Assessing the Impact of Reduced‐Precision Calculations On Solution Correctness S. Zhang et al. 10.1029/2019MS001817
- Bridging observations, theory and numerical simulation of the ocean using machine learning M. Sonnewald et al. 10.1088/1748-9326/ac0eb0
- Limits of reproducibility and hydrodynamic noise in atmospheric regional modelling B. Geyer et al. 10.1038/s43247-020-00085-4
- Mixed‐Precision for Linear Solvers in Global Geophysical Flows J. Ackmann et al. 10.1029/2022MS003148
- Weak constraint 4D-Var data assimilation in the Regional Ocean Modeling System (ROMS) using a saddle-point algorithm: Application to the California Current Circulation A. Moore et al. 10.1016/j.ocemod.2023.102262
- Number Formats, Error Mitigation, and Scope for 16‐Bit Arithmetics in Weather and Climate Modeling Analyzed With a Shallow Water Model M. Klöwer et al. 10.1029/2020MS002246
- swNEMO_v4.0: an ocean model based on NEMO4 for the new-generation Sunway supercomputer Y. Ye et al. 10.5194/gmd-15-5739-2022
- Fluid Simulations Accelerated With 16 Bits: Approaching 4x Speedup on A64FX by Squeezing ShallowWaters.jl Into Float16 M. Klöwer et al. 10.1029/2021MS002684
- Climate Modeling in Low Precision: Effects of Both Deterministic and Stochastic Rounding E. Paxton et al. 10.1175/JCLI-D-21-0343.1
- Characterizing uncertainties of Earth system modeling with heterogeneous many-core architecture computing Y. Yu et al. 10.5194/gmd-15-6695-2022
- Mixed precision algorithms in numerical linear algebra N. Higham & T. Mary 10.1017/S0962492922000022
16 citations as recorded by crossref.
- Compressing atmospheric data into its real information content M. Klöwer et al. 10.1038/s43588-021-00156-2
- Reduced-Precision Chemical Kinetics in Atmospheric Models K. Sophocleous & T. Christoudias 10.3390/atmos13091418
- Heterogeneous acceleration algorithms for shallow cumulus convection scheme over GPU clusters F. Li et al. 10.1016/j.future.2023.04.021
- Discussion on “Saving Storage in Climate Ensembles: A Model-Based Stochastic Approach” J. Bessac et al. 10.1007/s13253-023-00540-7
- Numerical algorithms for high-performance computational science J. Dongarra et al. 10.1098/rsta.2019.0066
- An Objective and Efficient Method for Assessing the Impact of Reduced‐Precision Calculations On Solution Correctness S. Zhang et al. 10.1029/2019MS001817
- Bridging observations, theory and numerical simulation of the ocean using machine learning M. Sonnewald et al. 10.1088/1748-9326/ac0eb0
- Limits of reproducibility and hydrodynamic noise in atmospheric regional modelling B. Geyer et al. 10.1038/s43247-020-00085-4
- Mixed‐Precision for Linear Solvers in Global Geophysical Flows J. Ackmann et al. 10.1029/2022MS003148
- Weak constraint 4D-Var data assimilation in the Regional Ocean Modeling System (ROMS) using a saddle-point algorithm: Application to the California Current Circulation A. Moore et al. 10.1016/j.ocemod.2023.102262
- Number Formats, Error Mitigation, and Scope for 16‐Bit Arithmetics in Weather and Climate Modeling Analyzed With a Shallow Water Model M. Klöwer et al. 10.1029/2020MS002246
- swNEMO_v4.0: an ocean model based on NEMO4 for the new-generation Sunway supercomputer Y. Ye et al. 10.5194/gmd-15-5739-2022
- Fluid Simulations Accelerated With 16 Bits: Approaching 4x Speedup on A64FX by Squeezing ShallowWaters.jl Into Float16 M. Klöwer et al. 10.1029/2021MS002684
- Climate Modeling in Low Precision: Effects of Both Deterministic and Stochastic Rounding E. Paxton et al. 10.1175/JCLI-D-21-0343.1
- Characterizing uncertainties of Earth system modeling with heterogeneous many-core architecture computing Y. Yu et al. 10.5194/gmd-15-6695-2022
- Mixed precision algorithms in numerical linear algebra N. Higham & T. Mary 10.1017/S0962492922000022
Latest update: 23 Apr 2024
Short summary
Mixed-precision approaches can provide substantial speed-ups for both computing- and memory-bound codes, requiring little effort. A novel method to enable modern and legacy codes to benefit from a reduction of precision without sacrificing accuracy is presented. Using a precision emulator and a divide-and-conquer algorithm it identifies the parts that cannot handle reduced precision and the ones that can. The method has been proved using two ocean models, NEMO and ROMS, with promising results.
Mixed-precision approaches can provide substantial speed-ups for both computing- and...
