Swiss Weather Service to Deploy World’s First GPU-Accelerated Supercomputer to Improve Weather Prediction
Update Time: 2019-12-20 13:58:18
NVIDIA today announced that the Swiss Federal Office of Meteorology and Climatology, MeteoSwiss, will be the first major national weather service to deploy a GPU-accelerated supercomputer to improve the accuracy of its daily weather forecasts.
The new system will provide more than two times higher resolution weather models at three times higher energy efficiency than MeteoSwiss’s existing system. Higher resolutions enable agencies like MeteoSwiss to predict storms and other weather hazards in more detail. This includes providing early warning forecasts for severe weather events, more precise tracking of the paths of dangerous storms and better insight into precipitation activity.
The higher resolution models will also enable MeteoSwiss to accurately model cloud formation and behavior, including predicting the amount, duration and location of rain, snow and related weather events in more detail. This capability is particularly important for Switzerland’s mountainous regions, where precipitation patterns are typically more difficult to forecast.
Powered by NVIDIA® Tesla® K80 GPU accelerators, the flagship offering of the Tesla Accelerated Computing Platform, the new MeteoSwiss weather prediction supercomputer provides 40 times higher performance than the CPU-based system it is replacing.
“Weather prediction is an incredibly difficult challenge, requiring massive amounts of computational horsepower to deliver accurate forecasts,” said Ian Buck, vice president of Accelerated Computing at NVIDIA. “The Tesla Accelerated Computing Platform is a natural fit for weather prediction and other large-scale, complex computational tasks, as it enables scientists and engineers to solve some of the world’s most difficult problems.”
The GPU-accelerated system will enable MeteoSwiss to produce two times higher resolution weather models for short-range forecasts (24 hour)1 , and three times higher resolution models for its medium-range forecasts2 .
It also will extend MeteoSwiss’s medium-range forecast capabilities with the addition of an ensemble prediction system with a forecast range of up to five days. Previously practical only for weather services with significantly larger infrastructure and resources, the ensemble prediction system uses high-performance GPUs to run multiple instances of computationally intensive simulations to provide information about the most likely evolution of the weather.
The new NVIDIA GPU-accelerated Cray weather prediction supercomputer has been installed at the Swiss National Supercomputing Centre (CSCS) in Lugano, Switzerland, and will be operational for weather forecasting by spring/summer 2016.
“The challenge with higher resolution weather models is the extremely short timeframe meteorologists have available to complete these simulations,” said Prof. Dr. Thomas Schulthess, director of CSCS. “GPU acceleration lets meteorologists quickly create high-resolution weather forecasts within this small window, giving them a considerably higher degree of confidence in the accuracy of their forecasts.”
Expanding Accessibility to GPU-accelerated COSMO Application
Daily weather forecasts are performed on high-performance supercomputers using numerical weather prediction models that simulate and predict the behavior of the atmosphere, including temperature, pressure, wind, precipitation and other factors.
MeteoSwiss produces its weather predictions using the COSMO model, which was developed by the Consortium for Small-scale Modeling. COSMO is also used extensively for operational weather forecasting by six other European national weather services in Germany, Italy, Greece, Poland, Romania and Russia.
The COSMO consortium is working to integrate MeteoSwiss’s GPU-accelerated developments into the standard version available for use by the other weather services in Europe and other members of the COSMO consortium.
About the Tesla Accelerated Computing Platform
The Tesla Accelerated Computing Platform is designed from the ground up for power-efficient, high performance computing, computational science, supercomputing, enterprise, data analytics and deep learning applications. It delivers dramatically higher performance and energy efficiency than a CPU-only approach. By deeply integrating the world’s fastest GPU accelerators, development tools, high-speed communication technology, a supported ecosystem and NVIDIA CUDA®, the world’s most pervasive parallel computing model, the Tesla platform delivers unprecedented application throughput in the data center.
Certain statements in this press release including, but not limited to, statements as to: the features, performance, benefits and availability of the NVIDIA Tesla K80 GPU-accelerated supercomputer for weather forecasting and its integration are forward-looking statements that are subject to risks and uncertainties that could cause results to be materially different than expectations. Important factors that could cause actual results to differ materially include: global economic conditions; our reliance on third parties to manufacture, assemble, package and test our products; the impact of technological development and competition; development of new products and technologies or enhancements to our existing product and technologies; market acceptance of our products or our partners' products; design, manufacturing or software defects; changes in consumer preferences or demands; changes in industry standards and interfaces; unexpected loss of performance of our products or technologies when integrated into systems; as well as other factors detailed from time to time in the reports NVIDIA files with the Securities and Exchange Commission, or SEC, including its Form 10-Q for the quarterly period ended July 26, 2015. Copies of reports filed with the SEC are posted on the company's website and are available from NVIDIA without charge. These forward-looking statements are not guarantees of future performance and speak only as of the date hereof, and, except as required by law, NVIDIA disclaims any obligation to update these forward-looking statements to reflect future events or circumstances.
GeForce GTX 860M VGA Graphic Chipset >
- 420GO V