2026· International Conference on Conceptual Structures· pp. 67-79· 0 citations· 13 references
Computer Science
TL;DR
This article analyzes the parallel scalability of the WILDFIRES-IGA-ADS simulator on a concurrent server with shared memory and employs the isogeometric finite element method with tensor product B-spline basis functions, explicit time integration schemes, and a linear computational cost solver resulting from the Kronecker product structure of the mass matrix.
We present BEAST, the first-ever Bayesian Swin Transformer for atmospheric forecasting on 0.25$^\circ$ global resolution able to accurately quantify both aleatoric and epistemic uncertainty. To overcome the associated computational bottlenecks, we devise an orthogonal 4D-parallelization scheme that introduces a unique domain-tensor-parallelism strategy and a novel uncertainty parallel method, enabling us to fully leverage GPU capacity and efficiently scale model training. For a 2.4-billion-parameter model, we achieve a peak performance of 3.96 EFLOP/s on 20,480 NVIDIA GH200 GPUs on the JUPITER supercomputer. We train BEAST as a 700-million-parameter model with 96 random weight samples on 384 nodes on 40 years of data for nearly one million gradient updates. This model achieves predictive skill scores competitive with state-of-the-art probabilistic atmospheric AI models and numerical models, and can predict extreme events with exceptional skill, while generating large ensembles 3 to 4 times faster than the current-best AI model. Our contribution unlocks the potential of high-fidelity uncertainty quantification in atmospheric AI models, heralding a new era for AI-based models in climate and Earth system sciences.
Deifilia Kieckhefen, J. P. G. H. Muriedas, L. Heyen et al.· 0 citations
We present FESOM2-JAX, a Python re-implementation of the Finite-volumE Sea ice-Ocean Model (FESOM2) in JAX. The model retains the unstructured-mesh, cell-vertex finite-volume formulation of the original, runs unchanged from a laptop CPU to 256 GPUs, and is end-to-end differentiable. FESOM2-JAX is a code shadow of the Fortran model: a projection onto the Python ecosystem, translated with large language models and verified kernel by kernel against the original. It is built to lower the barrier to experimentation, from new numerics and parameterizations to gradient-based calibration and hybrid physics-machine-learning components, while remaining close enough to the original so that what is developed in the shadow can be transferred back. In a 1958-2019 hindcast at 1$^{\circ}$ equivalent resolution with identical physics and forcing, the mean states of the JAX and Fortran versions differ from each other by two orders of magnitude less than either differs from observations, and the two runs agree for six decades in global temperature, salinity, heat content, and sea ice. The complete 1$^{\circ}$ configuration fits on a single GPU, a node of four GH200 superchips integrates $\sim$113 simulated years per wall-clock day, and meshes of up to 7.4 million surface vertices ($\sim$5 km) scale to 128 GPUs. What limits the model is communication rather than arithmetic. What the shadow adds to the original is the gradient: a single reverse-mode pass through the full time loop returns the sensitivity of a model diagnostic to a parameter at every mesh vertex, verified against finite differences. To our knowledge, FESOM2-JAX is the first global ocean-sea-ice model of CMIP-class complexity written natively in a differentiable framework, and the first on an unstructured mesh.
Nikolay Koldunov, S. Danilov, S. Cheedela et al.· 1 citation
The Galerkin triple product $A_c = P^T A P$ dominates the recurring per-solve setup cost of algebraic multigrid (AMG). For AMG on systems of PDEs the product is a rectangular-block sparse matrix triple product: for 3D elasticity the fine operator has $3\times3$ blocks, the prolongator $3\times6$, and the coarse operator $6\times6$, a shape no vendor sparse library supports. We map its algorithm space -- classical two-pass, fused-recompute, schedule-reordered, shared-memory-tiled, and inspector-executor variants -- under an explicit DRAM/L2 traffic model, and implement the leading variants in portable Kokkos (CUDA) and native CUDA backends using new PETSc blocked matrix types. Validated on an NVIDIA A100, the model predicts per level which variant moves the fewest bytes. Guided by it, a shared-memory-tiled kernel with a sorted, search-free schedule moves fewer bytes in less than half the time of the portable Kokkos team kernels on the fine-level product (10.5 vs 17.4 GB of DRAM, 45 vs 82 ms), within $2.5\times$ of the model's streaming floor for the full product and $1.9\times$ on its $A\cdot P$ stage. We further present prolongator filtering, a new PETSc GAMG algorithm that drops small blocks from the coarse space under a Frobenius criterion with a kernel-preserving projection; it reduces $P^TAP$ traffic, coarse-operator fill, and memory, and cuts the hot $P^TAP$ time $2.9\times$ on the fine grid with iteration counts unchanged. The driving application is a fully GPU-resident blocked pipeline in PETSc: finite-element assembly writes directly into the blocked device matrix, and the AMG setup, Galerkin products, and solve all operate on primary blocked data with no scalar expansion and no operator-sized device-host transfers in the recurring phases.
Process-based forest models are widely used to simulate forest growth and responses to environmental change, but their calibration and application often require many computationally expensive model evaluations. We present an implementation of the Physiological Processes Predicting Growth (3-PG) model in JAX that uses just-in-time compilation, vectorization, and GPU acceleration to reduce execution time. The implementation also supports automatic differentiation, providing gradients of model outputs and calibration objectives with respect to model parameters. This enables efficient gradient-based optimization and gradient-informed Bayesian calibration, extending 3-PG beyond conventional gradient-free approaches. The implementation produced results numerically consistent with r3PG for the evaluated configuration. Overall, the JAX implementation provides a faster and differentiable framework for calibrating and applying the 3-PG model.
G. M. Givi, Cédric Travelletti, Grégory Mermoud· 0 citations
A contiguous cartogram is a map whose area is proportional to a quantity such as population. The defining condition, that the Jacobian determinant of the deformation equals the density, is one equation for two unknown functions, so every cartogram method adds a tie-breaker, usually implicitly. Optimal transport makes the tie-breaker explicit: among all density-equalising maps of the frame onto itself, take the one that moves the population least in the mean-square sense. By Brenier's theorem that map is the gradient of a convex potential, so it has no local rotation anywhere and cannot fold. We compute this map for the world population of 2025 on a 4096 by 4096 Mercator grid (10 km cells) from the GHS-POP raster, using the fixed-point Monge-Amp\`ere iteration of Benamou, Froese and Oberman with spectral Poisson solves, continuation in the population share and an ocean-only buffer, on a laptop GPU. Against a Gastner-Newman diffusion cartogram of the same density we find the same transport cost to within 0.4 per cent and density errors of a few per cent for both, but a median local rotation of 8.8 degrees for diffusion against 0.01 for transport, and a median anisotropy of 6.30 against 3.93. The construction extends to local refinement, by transporting a city's 100 m population onto the area measure the global map assigns to it, and to a semi-discrete counterpart: 8,192 Laguerre cells of 1.00 million people each. Code, data provenance and every figure's inputs are public.
The lattice Boltzmann method (LBM) is a well-established approach for simulating fluid flows at the mesoscopic scale. With the stagnation of Moore's law, high-performance computing has shifted toward GPU accelerators, necessitating programming models that ensure both portability and efficiency across diverse hardware platforms. We present PortLBM, an extensible portable LBM framework built on SYCL that integrates cross-platform GPU support with interactive real-time visualization. PortLBM supports diverse simulation scenarios ranging from K\'arm\'an vortex streets and wing flows to porous media, and is designed for easy extension with new algorithms and backends. As part of a performance portability study, we evaluate PortLBM on contemporary GPU architectures from NVIDIA, AMD, and Intel, examining the impact of three data layouts (stream, bundle, and collision) and four algorithmic variants on simulation throughput. Our results show that no single configuration achieves optimal performance across all GPU vendors, confirming the need for system-specific tuning. The stream layout maximizes bandwidth and performs best on the contemporary NVIDIA and Intel GPUs, while the bundle layout improves cache efficiency and excels on the AMD GPU. Two-lattice schemes achieve higher throughput while one-lattice schemes are preferable under memory constraints. Our work underscores the necessity for adaptable, portable LBM software in increasingly heterogeneous computing environments.
Alexander Strack, Marcel Graf, Alexander Van Craen et al.· arXiv.org· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
