Seeking the best Weather Research and Forecasting model performance: an empirical score approach
- PDF / 1,246,670 Bytes
- 25 Pages / 439.37 x 666.142 pts Page_size
- 7 Downloads / 166 Views
Seeking the best Weather Research and Forecasting model performance: an empirical score approach R. Moreno1 · E. Arias1 · D. Cazorla1 · J. J. Pardo1 · F. J. Tapiador2
© Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract Weather forecasting, especially snowfall prediction, was critical in the 2018 Winter Olympics, where the accuracy of the predictions was of key importance for the planning of the different Olympic events. It was a significant challenge for the authors to meet the requirements in time and forecast resolution, while doing their best to be as competitive as possible. All the forecasts were obtained using the Weather Research and Forecasting (WRF) model, executed on the GALGO supercomputer. In order to obtain the best performance and meet the required execution times, different combinations of compilers, Message Passing Interface (MPI) libraries and computing platforms were tested to seek the best combinations. This work proposes an empirical score of special interest to supercomputer maintainers, developers and scientists, which can be useful to obtain the best WRF configuration for their systems. Additionally, we found substantial performance differences when using different combinations of compilers, MPI libraries and hybrid shared memory paradigms, although these differences varied depending on the underlying platform. As conclusion, after all the tests we performed, we chose the combination with Intel compilers, Intel MPI library and OpenMP for the production system tasked to perform the weather forecasts for the Winter Olympic Games. Keywords Optimization · Compilers · MPI · OpenMP · WRF
* R. Moreno [email protected] 1
University of Castilla-La Mancha, Albacete, Spain
2
University of Castilla-La Mancha, Toledo, Spain
13
Vol.:(0123456789)
R. Moreno et al.
1 Introduction Weather-related predictions and forecasts are of great use for an important international event such as the Olympic Winter Games. These games feature all types of winter sports, which require good weather conditions for the events to run smoothly. Knowing the upcoming weather conditions in advance is of special interest for the event organizers, for whom not only the amount of snowfall is important, but also the visibility level of a Ski Track or at the Olympic stadium. This kind of information is critical in order to plan or delay an event until conditions improve. The ICEPOP 2018 project emerged in order to provide abundant meteorological data during the Winter Games from different sources around the world. The authors of this work faced many challenges when they collaborated by feeding data to the ICE-POP 2018 project. The Weather Research and Forecasting (WRF) model, which is a Numerical Weather Prediction (NWP) model, was used to provide 24-h-ahead forecasts for the different stations and points of interest of the Winter Games, with a 12-h throughput frequency. WRF physics integration was performed over three nested grids, or domains, covering a target geographical of the Oly
Data Loading...
