Kilometer-scale Earth system models (ESMs) necessitate exascale supercomputers to facilitate realistic simulations of weather phenomena and climate variability over a time span ranging from days to decades. We present AP3ESM, an ultra‑high‑resolution, AI‑Powered, Performance‑Portable ESM coupling atmosphere, land surface, ocean, and sea ice components. By leveraging the performance portability features of Kokkos and OpenMP, the AP3ESM operates efficiently on two heterogeneous systems while incurring minimal development overhead. Advanced optimization techniques, such as adaptive parallel algorithms, AI-enhanced physical parameterizations, and mixed-precision computations, have been implemented to further boost the computational efficiency. Breaking the 1-km resolution barrier, AP3ESM delivers 0.85 and 1.98 simulated-years-per-day (SYPD) for the standalone atmosphere and ocean components on 34.1 million Sunway cores and 16085 GPUs, respectively; the holistic AP3ESM achieves 0.54 SYPD on 37.2 million Sunway cores. Notably, the forecast experiment successfully captures Super Typhoon Doksuri in 2023 and its associated extreme rainfall across China.

Sea surface temperature (SST) is critically important for understanding ocean dynamics and supporting various marine activities, making accurate short-term SST forecasting highly significant. However, accurately modeling the multi-scale variability of SST remains challenging for existing deep learning (DL) models. This study introduces the Coupled Transformer-CNN Network (CoTCN), a hybrid architecture designed to leverage the multi-scale variability of SST. The CoTCN combines the strengths of Transformers and convolutional neural networks (CNNs), significantly enhancing SST forecasts’ spatial continuity and predictive accuracy. Compared to five state-of-the-art DL models based on Transformer or CNN that include ConvLSTM, ConvGRU, AFNO, PredRNN, and SwinLSTM, CoTCN demonstrates superior performance in global and local areas of SST forecasting. At 1-day lead time, CoTCN reduces the global average root mean square error (RMSE) by over 15%, with forecast errors ranging from 0.20°C to 0.53°C across 1–10 day lead times. Moreover, the CoTCN effectively mitigates the checkerboard artifacts inherent to the Vision Transformer architecture. These findings highlight the effectiveness of CoTCN in capturing SST’s multi-scale features and underscore the promising potential of hybrid architectures for future DL models.

It is agreed that portfolio selection is of great importance for the financial market. Numerous outstanding exact and heuristic algorithms have been proposed in the past decades. However, their development always demands meticulous human ideas and could be time-consuming. Moreover, most of them tend to suffer from performance degradation when exposed to new portfolio selection models and different investment environments. Learning-enabled approaches have recently yielded impressive results, but these methods still grapple with challenges in model design and training. In this paper, we explore the mutual facilitation of large language models (LLMs) and huristic approaches in portfolio selection, and propose a novel LLM-based multi-objective evolutionary algorithm (MOEA) named IlmPC-NSGA-II. In this algorithm, the LLM with carefully-designed well-structured prompts serves as a straightforward yet effective engine for generating new solutions, non-dominated sorting and crowding distance calculation are adopted to enable the LLM and the evolutionary process to mutually guide toward the optimal region of the solution space. Experimental results on various scales of constrained multi-objective portfolio selection models and four benchmark problems demonstrate that our proposed approach can achieve a more competitive performance compared to widely-used MOEAs and the LLM-only method.

Ocean general circulation models (OGCMs) are indispensable for studying the multi-scale oceanic processes and climate change. High-resolution ocean simulations require immense computational power and thus become a challenge in climate science. We present LICOMK++, a performance-portable OGCM using Kokkos, to facilitate global kilometer-scale ocean simulations. The breakthroughs include: (1) we enhance cuttingedge Kokkos with the Sunway architecture, enabling LICOMK++ to become the first performance-portable OGCM on diversified architectures, i.e., Sunway processors, CUDA/HIP-based GPUs, and ARM CPUs. (2) LICOMK++ overcomes the one simulated-years-per-day (SYPD) performance challenge for global realistic OGCM at $1-\mathrm{km}$ resolution. It records $\mathbf{1. 0 5}$ and 1.70 SYPD with a parallel efficiency of 54.8% and 55.6% scaling on almost the entire new Sunway supercomputer and two-thirds of the ORISE supercomputer. (3) LICOMK++ is the first global 1-km-resolution realistic OGCM to generate scientific results. It successfully reproduces mesoscale and submesoscale structures that have considerable climate effects.

Recent works in deep learning have shown that large models can dramatically improve performance. In this paper, we accelerated the deep network training using many GPUs. We have developed a framework based on Caffe called Caffe-HPC that can utilize computing clusters with multiple GPUs to train large models. Caffe[6] provides multimedia scientists and practitioners with a clean and modifiable framework for state-of-the-art deep learning algorithms and a collection of reference models. And Caffe-HPC retains all the features of the original Caffe, the model trained on original Caffe can be continue to trained on Caffe-HPC. It provides a convenient solution for people who are using Caffe and want to speed up the training. Using an Asynchronous Stochastic Gradient Descent optimizer, We made a good acceleration on training a CNN model on ILSVRC[5] 2012 dataset. And we have compared the convergence of different SGD algorithms. We believe our work will makes it possible to train larger networks on larger training sets in a reasonable amount of time.

Summary Load imbalance is a common problem to be tackled urgently in large scale data‐driven simulation systems or data intensive computing. According to the coupler, the Chinese Academy of Sciences‐Earth System Model (CAS‐ESM) implements one‐way nesting of the Institute of Atmospheric Physics of Chinese Academy of Sciences Atmospheric General Circulation Model version 4.0 (IAP AGCM4.0) and Weather Research and Forecasting model (WRF). The METGRID (meteorological grid) and REAL program modules in the WRF are used to process meteorological data. In the CAS‐ESM, the load of the METGRID module is seriously unbalanced on many CPU cores. The load imbalance has a serious impact on the processing speed of meteorological data, so this study designs an optimization algorithm to solve the problem. Numerical experiments show that compared to before optimization, the optimization algorithm can solve the load imbalance of the METGRID, and the computation speed of the METGRID and REAL modules after optimization on 64 CPU cores is about 7.2 times faster than before. Meanwhile, the whole computation speed of the CAS‐ESM can improve by 217.53%. In addition, results indicate that they also can reach to a similar speedup on different numbers of CPU cores. Copyright © 2016 John Wiley & Sons, Ltd.

Abstract Precipitation forecast verification is essential to the quality of a forecast. The Gaussian mixture model (GMM) can be used to approximate the precipitation of several rain bands and provide a concise view of the data, which is especially useful for comparing forecast and observation data. The robustness of such comparison mainly depends on the consistency of and the correspondence between the extracted rain bands in the forecast and observation data. We propose a novel co‐estimation approach based on GMM in which forecast and observation data are analysed simultaneously. This approach naturally increases the consistency of and correspondence between the extracted rain bands by exploiting the similarity between both forecast and observation data. Moreover, a novel visualization and exploration framework is implemented to help the meteorologists gain insight from the forecast. The proposed approach was applied to the forecast and observation data provided by the China Meteorological Administration. The results are evaluated by meteorologists and novel insight has been gained.

This paper proposed a parallel refined block Arnoldi method for computing a few eigenvalues with largest or smallest real parts. The method accelerated by Chebyshev iteration is also investigated. We report some numerical results and compare the parallel refined block methods with single vector counterparts. The results show that the proposed method is more efficient than single vector counterparts.