The National Oceanic and Atmospheric Administration (NOAA) has selected Google Cloud as the primary provider of high-performance computing (HPC) infrastructure for its Weather and Climate Operational Supercomputing System (WCOSS). The move marks a major step in the agency's modernization strategy and positions NOAA among the first operational numerical weather prediction centers to shift core forecasting operations to a public cloud platform.

NOAA said the transition is designed to improve the speed, flexibility, and reliability of weather forecasting models used by the National Weather Service. The agency plans to begin migrating key operational models, including the Global Forecast System (GFS) and Global Ensemble Forecast System (GEFS), in early 2027. The migration is expected to be completed by December 2027.

Cloud Infrastructure Aims To Improve Forecast Performance

Google Cloud's H4D virtual machines, powered by fifth-generation AMD EPYC processors, will provide the computing platform for NOAA's weather simulations. These systems are designed for large-scale numerical weather prediction that requires thousands of processors to work together with low-latency communication.

According to NOAA Administrator Neil Jacobs, cloud-based HPC will reduce limitations associated with traditional on-premises systems. The agency expects faster model development, improved resilience, and the ability to scale computing resources during periods of high demand, such as tropical cyclone season. NOAA National Weather Service Director Ken Graham said the modernization will help operational models adapt more quickly to scientific advances while supporting the agency's mission to protect life and property.

Long-Term Partnership Expands

The announcement builds on a long-standing collaboration between NOAA and Google Cloud. Previous projects have included hosting environmental datasets, supporting wildfire monitoring, and developing cloud-based computing tools for scientific research. Google also highlighted earlier work involving advanced weather prediction research and artificial intelligence models.

The transition reflects a broader trend among meteorological agencies worldwide to increase computing capacity as weather models become more complex. More powerful computing enables higher-resolution forecasts and faster processing of growing volumes of satellite, radar, and observational data. Experts say continued investment in computing infrastructure remains critical as extreme weather events become more frequent and forecasting demands increase.