论文标题
Fourcastnet：使用自适应傅立叶神经操作员加速全球高分辨率天气预报
FourCastNet: Accelerating Global High-Resolution Weather Forecasting using Adaptive Fourier Neural Operators
论文作者
论文摘要
气候变化所扩大的极端天气正在造成全球日益毁灭性的影响。由于高计算成本和严格的时间到解决方案限制，目前使用基于物理的数值天气预测（NWP）的使用限制了精度。我们报告说，数据驱动的深度学习地球系统模拟器Fourcastnet可以预测全球天气，并在接近最先进的准确性的同时，比NWP更快地产生五个量子命令。四个超级计算系统有效地进行了优化，并在三个超级计算系统上进行了有效尺度：Selene，Perlmutter和Juwels Booster高达3,808个NVIDIA A100 GPU，在混合精度中达到140.8 PETAFLOPS（该峰值为11.9％）。在3,072GPU上测量的训练四界训练四界的训练时间为67.4分钟，相对于最先进的NWP，在推理中，相对于最先进的NWP的时间更快。 Fourcastnet提前一周可产生准确的瞬时天气预测，使得可以更好地捕捉天气的巨大合奏，并支持更高的全球预测决议。
Extreme weather amplified by climate change is causing increasingly devastating impacts across the globe. The current use of physics-based numerical weather prediction (NWP) limits accuracy due to high computational cost and strict time-to-solution limits. We report that a data-driven deep learning Earth system emulator, FourCastNet, can predict global weather and generate medium-range forecasts five orders-of-magnitude faster than NWP while approaching state-of-the-art accuracy. FourCast-Net is optimized and scales efficiently on three supercomputing systems: Selene, Perlmutter, and JUWELS Booster up to 3,808 NVIDIA A100 GPUs, attaining 140.8 petaFLOPS in mixed precision (11.9%of peak at that scale). The time-to-solution for training FourCastNet measured on JUWELS Booster on 3,072GPUs is 67.4minutes, resulting in an 80,000times faster time-to-solution relative to state-of-the-art NWP, in inference. FourCastNet produces accurate instantaneous weather predictions for a week in advance, enables enormous ensembles that better capture weather extremes, and supports higher global forecast resolutions.
