Supercomputer maker Cray will get $36 million to upgrade the systems running a European weather forecasting facility, highlighting the growing use of such massive infrastructures by meteorological and climate research facilities.
Cray officials announced the agreement Jan. 25, saying the company will upgrade the supercomputers and high-end storage systems that it put in place for the European Centre for Medium-Range Weather Forecasts almost three years ago. The ability to easily upgrade its systems is a key factor in the design of Cray’s supercomputers, according to Catalin Morosanu, the company’s vice president of sales for the Europe, Middle East and Africa (EMEA) region.
“This allows our customers to have a lower total cost of ownership of their systems over time, and quickly be able to deploy the latest and most-advanced technologies,” Morosanu said in a statement.
In June 2013, Cray was awarded a contract by the European facility to deliver two of its XC30 supercomputers and one of its Sonexion storage systems. Now the supercomputer maker will upgrade the facility with its XC40 systems—which will include the next-generation Xeon processors from Intel—and another Sonexion 2000 scale-out Lustre storage system. In addition, Cray will provide the center with a 32-node XC40-AC system, which will be powered by the latest generation of Intel’s Xeon Phi many-core processors, dubbed “Knights Landing.”
Cray will deliver the new systems later this year.
Intel officially unveiled the 14-nanometer Knights Landing processor at the SC 15 supercomputer show in November 2015. The chip maker had rolled out the first of the Xeon Phis in 2011, a 22nm 60-core chip that was designed to be used as a coprocessor that works alongside primary Xeon processors, playing a role similar to the GPU accelerators offered by Nvidia and Advanced Micro Devices.
With Knights Landing, Intel is offering a chip with as many as 72 cores and that can be used as either a coprocessor or primary processor. It was built on the company’s Silvermont architecture and has more than 8 billion transistors and will deliver more than 3 teraflops of peak performance, according to officials.
Cray introduced the XC40 supercomputer in September 2014. The system includes the Aries system interconnect technology, a network topology dubbed Dragonfly and, as an option, Cray’s DataWarp I/O accelerator technology that leverages solid-state drives (SSDs), which officials at the time said delivers up to five times the performance of spinning disks at a similar cost. The XC40 also includes a liquid-cooling technology and a tightly integrated Cray Linux Environment software offering.
It also can use an array of processor types, including Intel’s Xeon and Xeon Phi chips and Nvidia’s Tesla GPU accelerators.
The XC40-AC is a less-dense version that is air-cooled.
The European Centre for Medium-Range Weather Forecasts is based in the United Kingdom and supported by 34 countries. It creates global models and data assimilation systems and specializes in global weather predictions up to a few weeks in advance as well as longer-range forecasts for up to a year in the future. It also runs the Copernicus Atmosphere Monitoring and Climate Change Services, run by the European Union.
The upgraded supercomputers will enable the facility to improve its modeling and predictions of severe weather events in a two-week range, officials said.
Cray’s announcement is the latest in a series from supercomputer makers about new and enhanced systems being used by weather forecasters and climate researchers. Most recently, SGI earlier this month announced it is building the next-generation system for the National Center for Atmospheric Research that will be based on the vendor’s ICE XA system and will be 2.5 times more powerful and three times more energy-efficient than the facility’s current system.
Both Cray and SGI over the past couple of years have made other announcements around the work they were doing with such facilities as NASA’s Center for Climate Simulation, the National Oceanic and Atmospheric Administration (NOAA), Australian Bureau of Meteorology, Swiss National Supercomputing Centre and MeteoSwiss, and Finnish Meteorological Institute.
In a post on his company’s blog in September 2015, Philip Brown, earth sciences segment leader at Cray, wrote that “improving and expanding weather and climate forecasting activities is a challenging proposition, requiring investment in fundamental scientific research, development of more scalable numerical models, and enhanced supercomputer resources upon which those models can run. But the benefits are clear: In addition to the expansion of scientific knowledge, investment in weather and climate forecasting provides important socio-economic returns.”