Beyond Design: Skip-layer Routing—The Waveguide Structure That Makes 224G Possible
Designers Notebook: Old vs. New School System-level Packaging—Flip-Chip to Chiplets
Elementary, Mr. Watson: Builders of the Dream
HPE announced that Earth Sciences New Zealand (formally NIWA) of Aotearoa (New Zealand) selected HPE Cray XD2000, purpose-built for AI and simulation workloads, to accelerate the organization’s environmental science and precision of meteorological forecasting.
The new system, named Cascade, replaces Earth Sciences New Zealand’s aging high performance computing (HPC) and delivers a threefold increase in computational power for researchers to run multiple AI-powered simulations simultaneously for more accurate weather predictions. The new system supports Earth Sciences New Zealand’s mission to strengthen the country’s resilience against weather and climate-related hazards, such as wildfires and flooding from heavy rainfall.
Cascade is powered by AMD 4th Gen EPYC processors and tightly coupled with HPE GreenLake for File Storage to deliver a highly performant, simplified storage environment and enable 19 petabytes of data to accelerate local research, supporting faster decision-making and more informed crisis management.
The Cascade system features direct liquid cooling (DLC) technology that minimizes energy use and water usage and efficiently cools the computational cluster, which was architected flexibly to allow future expansion. Furthermore, the electricity supplied to the CDC data centre, where Cascade is housed, is generated from 100% renewable sources.
“We’re thrilled to partner with Earth Sciences New Zealand on this groundbreaking initiative,” said Chris Weber, Vice President and Managing Director, South Pacific at HPE. “By delivering advanced HPC and AI capabilities, we’re empowering New Zealand’s leading climate and environmental researchers with the tools they need to drive scientific discovery and tackle some of the most pressing challenges of our time. Importantly, we’re able to maximise energy efficiency through direct liquid cooling and leverage renewable energy. This collaboration showcases the true potential of technology to make a meaningful impact, both locally and globally.”
In another milestone, Earth Sciences New Zealand’s Cascade is set to be the largest implementation in the Asia Pacific region of HPE GreenLake for File Storage, across both HPC and traditional enterprise computing. For Earth Sciences New Zealand, it represents a significant commitment to future-proof the country against the impacts of climate change and ensuring public safety for generations to come.
Based in Auckland, Earth Sciences New Zealand is the country’s leading provider of natural and environmental science services, internationally recognized for its cutting-edge research in climate, freshwater, geological resources, energy futures and marine science. As an island nation, New Zealand faces unique challenges from atmospheric rivers – narrow corridors in the atmosphere that carry large quantities of water vapor and are known to cause major rainfall events – which makes Earth Sciences New Zealand’s role in forecasting weather even more vital to support its national climate resilience.
“This upgrade with HPE Cray XD2000 marks a huge leap forward in our ability to understand and predict the environment around us,” says Dr. Jess Robertson, Chief Scientist – Advanced Technology at Earth Sciences New Zealand. “Enhancing scientific research with the advanced technology solutions gives us the tools to make smarter and safer decisions with confidence for New Zealand in a changing climate.”
Alongside its forecasting capabilities, Earth Sciences New Zealand also supports the country’s low-carbon economy and biodiversity protection, improving water quality, social equity, and regional development.
Partnering with HPE, Earth Sciences New Zealand is able to achieve unprecedented performance for its workloads, supercharging research data, processing and forecasting capabilities. Leveraging HPE’s extensive design and liquid cooling expertise, Cascade is purpose-built to operate at full capacity for longer, maximizing scalability and up-time. Cascade will also use real-time data to unlock an advanced weather forecasting technique referred to as nowcasting. Nowcasting uses AI to predict short-term, highly-localized weather rapidly, supporting the inclusion of solar and wind in high-frequency energy markets as well as emergency response to fires and extreme weather.
