HPE is advancing weather and climate research with breakthrough supercomputing systems at any scale, delivering the robust performance to help protect and save lives.
Researchers rely on high compute performance, increasingly combined with artificial intelligence, for accurate daily weather predictions, warnings of severe weather, and climate monitoring. To manage these enormous workloads, weather organizations are searching for a new breed of supercomputing systems.
Severe weather can have catastrophic effects, even in communities with developed infrastructure. In 2021, the U.S. alone has recorded 18 weather or climate disasters with losses exceeding $1 billion per event. The disaster costs for the first nine months of this year total $104.8 billion, already surpassing the costs for all of 2020. According to NOAA, the number and cost of disasters are increasing over time due to a combination of increased exposure (valuables at risk of loss), vulnerability (how much damage can occur in a specific location), and the effects of climate change.
Todayโs researchers are working around the clock to keep up with the increasing intensity and frequency of extreme events. One key challenge is understanding the weather in real time.
The accuracy of weather models is limited by the type of data used to initialize them and by how quickly the data can be processed. Supercomputers make it possible to leverage data from numerous sources and process highly complex workloads almost instantly. The greater the compute, the more accurate a model can become with fewer approximations, so weather organizations can make forecasts with increased precision while lowering the expense of running slower models.
As weather continues to change, data becomes perishable in seconds. To predict the next major event, weather organizations are searching for supercomputing technologies that can meet their unique requirements, regardless of the scope or types of workloads they run. Right-sized supercomputing systems are the key to improving the validity of weather models to make timely predictions and deliver faster, more accurate forecasts.
Next-generation compute for advanced weather forecasting
From one of the most effective supercomputers, HPE has developed an entirely new solution that enhances and accelerates weather research. These breakthrough systems are engineered to address the rising demands on weather organizations, providing the robust performance needed to produce critical insights when and where they are needed most. With over 40 years of experience in the industry and systems at many of the worldโs weather centers, HPE helps organizations of every size and scope create an environment built for growth and innovation. Now, weather teams can tackle compute and scientific challenges with ease as data proliferates, core counts increase, and workflows expand.
To accomplish these goals, HPE is bringing the capabilities of our landmark HPE Cray EX supercomputers to the enterprise. The robust portfolio features industry-leading supercomputing hardware and software designed for the most demanding workloads, making it a supercomputer of choice across the world. With the capacity to build out, add features, and upgrade infrastructure as needed, organizations can use achieve accurate and timely insights from increasingly data-intensive weather models.
Top weather organizations are deploying these high-performance systems to bring new possibilities to weather prediction and accelerate forecasts that will help communities prepare for natural disasters. The Met Office in the UK has announced a new 60-petaflops supercomputer built on an HPE Cray EX system that will provide more accurate weather prediction and climate change modeling using AI, modeling, and simulation with the scalability of the cloud. Their goal is to use this machine to help build a more resilient world in a changing climate.
Now, HPE has introduced the HPE Cray EX2500, a smaller version of the HPE Cray EX that allows weather organizations to harness supercomputing capabilities on a smaller scale, at a price they can afford. By implementing the same components as the massive systems, organizations can perform rapid modeling and analysis while matching data with other organizations to improve collaboration and multidisciplinary insights that will help protect and save lives.
HPE Cray EX2500 supercomputers feature a compact system architecture with liquid cooling that enables high efficiency without impacting performance. As workloads grow in complexity demanding higher speed and more compute, the systems continue to provide significant cost savings. Direct liquid cooling optimizes price-performance at any scale, so weather organizations can utilize the right level of compute within their specific budgets, power sourcing, and space requirements.
HPE Cray EX server blades are the foundation for the new right-sized systems, which enable accelerator-based computing with two nodes per blade, a dense configuration that reduces overall power consumption and operational costs.
Server blades are powered by a blend of CPUs and accelerators, which gives weather organizations the flexibility to optimize their infrastructure for different workflows.
To streamline the communication between servers, supercomputing systems from HPE leverage HPE Slingshot interconnect for consistent, reliable throughput with efficient congestion management.
Additionally, we offer world-leading software, accelerating time to results, reproducibility and resilience, including HPE Performance Cluster Manager system management software to keep the systems healthy and running at full capacity so they can be available to different users from anywhere, at any time. HPE Cray Programming Environment is a complete toolchain which helps researchers develop and tune even the most complex weather codes, so they run fast regardless underlying architecture.
On the storage side, purpose-built HPC storage solutions enable data to surge through this high-speed network, delivering vast amounts of information to the hands of researchers. These technologies are designed to handle the demands of supercomputing environments with unmatched agility, even as severe weather events heighten the need for data processing.
For help getting started, you get proven expertise and support services through HPE Pointnext Services and cloud services from HPE GreenLake. These offerings are tailored to help organizations identify their goals for supercomputing and create a roadmap to success, whether selecting the right supercomputing technologies or allowing weather organizations to pay only for what they use.
At HPE we're committed to pioneering a new age in weather forecasting and climate research. We have a long-term roadmap to help companies benefit from the most advanced and cost-effective computational elements. This is your time to innovate.
Contact us today to make the next leap in supercomputing.
Advantage EX Experts
Hewlett Packard Enterprise
twitter.com/hpe_hpc
linkedin.com/showcase/hpe-ai/
hpe.com/info/hpc
AdvEXperts
Our team of Hewlett Packard Enterprise Advantage EX experts helps you dive deep into high performance computing and supercomputing topics.
