The ESCAPE project: Energy-efficient Scalable Algorithms for Weather Prediction at Exascale
العنوان: | The ESCAPE project: Energy-efficient Scalable Algorithms for Weather Prediction at Exascale |
---|---|
المؤلفون: | Milosz Ciznicki, Bent Hansen Sass, Geert Smet, Sami Saarinen, Christian Kühnlein, Gianmarco Mengaldo, Valentin Clément, Xavier Vigouroux, Michal Kulczewski, Michael Glinton, Carlos Osuna, Nils Wedi, Willem Deconinck, Alan Gray, Mats Hamrud, Philippe Marguinaud, Pierre Bénard, Alexander J. Macfaden, Nick New, Parijat Shukla, David Guibert, Erwan Raffin, Bartosz Bosak, George Mozdzynski, Fabrice Voitus, Joanna Szmelter, Yongjun Zheng, Sarah-Jane Lock, Alastair McKinstry, Cyril Mazauric, Charles Colavolpe, Marcin Procyk, Per Berg, Louis Douriez, Michael Lysaght, Sebastian Ciesielski, Mike Gillard, Pawel Spychala, Michail Diamantakis, Michael Lange, Kristian Pagh Nielsen, Jacob Weismann Poulsen, Wojciech Piątek, Piotr K. Smolarkiewicz, Krzysztof Kurowski, Peter Messmer, Oisín Robinson, Daniel Thiemert, Oliver Fuhrer, Andreas Müller, Piet Termonia, Peter Bauer, Daan Degrauwe, Andrzej A. Wyszogrodzki, Joris van Bever, Enda O’Brien, Michael Baldauf, Zbigniew P. Piotrowski, Marek Błażewicz |
المصدر: | Geoscientific Model Development, Vol 12, Pp 4425-4441 (2019) Geoscientific Model Development Discussions GEOSCIENTIFIC MODEL DEVELOPMENT |
بيانات النشر: | Copernicus GmbH, 2019. |
سنة النشر: | 2019 |
مصطلحات موضوعية: | 010504 meteorology & atmospheric sciences, Computer science, Distributed computing, Weather and climate, 02 engineering and technology, 01 natural sciences, 7. Clean energy, Software portability, 0202 electrical engineering, electronic engineering, information engineering, 0105 earth and related environmental sciences, 020203 distributed computing, lcsh:QE1-996.5, Solver, PERFORMANCE, Numerical weather prediction, Supercomputer, CLIMATE, lcsh:Geology, 13. Climate action, Earth and Environmental Sciences, SIMULATION, Programming paradigm, ATMOSPHERIC MODELS, Xeon Phi, Efficient energy use, GENERATION |
الوصف: | In the simulation of complex multi-scale flows arising in weather and climate modelling, one of the biggest challenges is to satisfy strict service requirements in terms of time to solution and to satisfy budgetary constraints in terms of energy to solution, without compromising the accuracy and stability of the application. These simulations require algorithms that minimise the energy footprint along with the time required to produce a solution, maintain the physically required level of accuracy, are numerically stable, and are resilient in case of hardware failure. The European Centre for Medium-Range Weather Forecasts (ECMWF) led the ESCAPE (Energy-efficient Scalable Algorithms for Weather Prediction at Exascale) project, funded by Horizon 2020 (H2020) under the FET-HPC (Future and Emerging Technologies in High Performance Computing) initiative. The goal of ESCAPE was to develop a sustainable strategy to evolve weather and climate prediction models to next-generation computing technologies. The project partners incorporate the expertise of leading European regional forecasting consortia, university research, experienced high-performance computing centres, and hardware vendors. This paper presents an overview of the ESCAPE strategy: (i) identify domain-specific key algorithmic motifs in weather prediction and climate models (which we term Weather & Climate Dwarfs), (ii) categorise them in terms of computational and communication patterns while (iii) adapting them to different hardware architectures with alternative programming models, (iv) analyse the challenges in optimising, and (v) find alternative algorithms for the same scheme. The participating weather prediction models are the following: IFS (Integrated Forecasting System); ALARO, a combination of AROME (Application de la Recherche à l'Opérationnel à Meso-Echelle) and ALADIN (Aire Limitée Adaptation Dynamique Développement International); and COSMO–EULAG, a combination of COSMO (Consortium for Small-scale Modeling) and EULAG (Eulerian and semi-Lagrangian fluid solver). For many of the weather and climate dwarfs ESCAPE provides prototype implementations on different hardware architectures (mainly Intel Skylake CPUs, NVIDIA GPUs, Intel Xeon Phi, Optalysys optical processor) with different programming models. The spectral transform dwarf represents a detailed example of the co-design cycle of an ESCAPE dwarf. The dwarf concept has proven to be extremely useful for the rapid prototyping of alternative algorithms and their interaction with hardware; e.g. the use of a domain-specific language (DSL). Manual adaptations have led to substantial accelerations of key algorithms in numerical weather prediction (NWP) but are not a general recipe for the performance portability of complex NWP models. Existing DSLs are found to require further evolution but are promising tools for achieving the latter. Measurements of energy and time to solution suggest that a future focus needs to be on exploiting the simultaneous use of all available resources in hybrid CPU–GPU arrangements. |
وصف الملف: | application/pdf |
تدمد: | 1991-959X 1991-9603 |
URL الوصول: | https://explore.openaire.eu/search/publication?articleId=doi_dedup___::83025fb17a365f9a6ac51515f1206d0b https://doi.org/10.5194/gmd-2018-304 |
حقوق: | OPEN |
رقم الأكسشن: | edsair.doi.dedup.....83025fb17a365f9a6ac51515f1206d0b |
قاعدة البيانات: | OpenAIRE |
تدمد: | 1991959X 19919603 |
---|