Image credit: Stock
US wildfire science firm Technosylva has announced the launch of what it is calling the world’s largest dedicated supercomputers for wildfire modelling.
The system, developed together with PSSC Labs, simulates more than one billion fire scenarios each day, applying artificial intelligence to decades of fire weather data to identify communities and electric lines most at risk.
According to Technosylva, results are delivered in seconds providing utilities and fire agencies precious lead time to protect infrastructure and lives.
Joaquin Ramirez, chief technology officer and founder of Technosylva, said: “Accurate and timely wildfire forecasting requires supercomputing at scale.”
The 11,500-core system incorporates more data than what standard, publicly available weather and fuel models provide, including live and dead fuel moisture, vegetation conditions, terrain-driven winds, and three decades of historical fire weather patterns.
Have you read?
UK grid is resilient to Iberia-style blackout says NESO report
US looks to speed up transmission and generation projects for AI
Forecasts are generated at two-kilometer resolution and draw on terabytes of data – thousands of critical fire variable data feeds – processed every few hours.
The system runs many wildfire models in parallel and in conjunction, including refined, proprietary models for fire spread, crown spotting, extreme weather, fuel moisture and urban conflagration.
That scale allows utilities and agencies to make precise power shutoff decisions and accelerate evacuation alerts, states Technosylva.
Ramirez added: “This platform fuses high-performance computing with advanced AI, running ensemble forecasts against decades of meteorological and fire-behaviour data.
“The result is faster, more precise predictions that set a new benchmark for operational wildfire readiness.”
According to the US Forest Service, wildfires have destroyed more than 35,000 structures over the past decade, making high-performance computing and AI critical to understanding and mitigating the complexity of modern wildfire behavior.
Brent Shaw, senior numerical weather prediction architect at Technosylva, said: “We can process massive datasets in hours and deliver forecasts in seconds. It is like upgrading from an X-ray to a real-time MRI of wildfire behavior across the entire country.
“The same system also makes it possible to train and deploy AI models that capture wildfire dynamics in real-time.”
