Papers for

weather prediction centers

Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.

GPU compression cuts climate model communication time greatly

Toward GPU-Resident Climate Models: A Feasibility Study on Lossy Compression for the Spherical Harmonic Transform's Communication Bottleneck

Abstract: Operational pseudospectral atmospheric models such as the ECMWF Integrated Forecasting System (IFS) run today almost exclusively on CPUs; GPU ports are under active development but not yet used in production. These models rely on the Spherical Harmonic Transform (SHT). Each time-step requires forward and inverse SHTs, and both passes depend on global pencil transposition that redistribute multi-dimensional arrays across compute nodes. At large node counts these global collectives dominate wall-clock time. We investigate GPU-resident lossy compression, using representative fields from the DYAMOND high-resolution operational dataset as input, and combining measured GPU compression throughput with SimGrid network simulation, we show that ZFP at 16 bits per value (rate-16), about the same storage budget as float16, matches the communication-time reduction of float16 truncation while delivering approximately $\mathbf{1600\times}$ lower mean relative error. ZFP at 8 bits per value (rate-8) achieves approximately $\mathbf{1.93\times}$ the speedup of float16 while retaining $\mathbf{4\times}$ lower mean relative error.

Mon 21 SeptDistributed, Parallel, and Cluster Computing
The gist
Climate models that predict weather and atmosphere behavior usually run on CPUs, but switching them to GPUs could speed things up. The models rely on a math step called the Spherical Harmonic Transform, which needs large data exchanges that slow everything down. The authors tested a special lossy compression method on GPUs that shrinks the data exchanged without losing much accuracy. Their approach reduces communication time significantly more than simple truncation and keeps errors much lower, making GPU-based climate models more practical.
Open 2609.24294v1