Video: Cray Powers More Accurate Forecasts at ECMWF

Print Friendly, PDF & Email


In this video, Al Roker from the Today Show looks at how Cray XC30 supercomputers give ECMWF more accurate forecasts than we get here in America. ECMWF uses advanced computer modeling techniques to analyze observations and predict future weather. Their assimilation system uses 40 million observations a day from more than 50 different instruments on satellites, and from many ground-based and airborne measurement systems.

“ECMWF’s multi-petaflops supercomputer facility is designed for operational resiliency featuring two Cray XC30 systems and independent Cray Sonexion storage systems. The system comprises two independent subsystems located in separate halls. It has separate resilient power and cooling systems to protect against a wide range of possible failures. Each subsystem consists of 19 Cray XC30 cabinets equipped with Intel Ivy Bridge processors and around 3500 dual-socket compute nodes per system, a number of Cray Development and Login nodes and more than 6 petabytes of Lustre storage with the ability to cross mount the Lustre file systems between the halls.”

ECMWF’s supercomputers are among the largest of their type in Europe:

  • Operate with a sustained speed of more than 200 trillion floating point operations per second.
  • Our supercomputers serve a variety of purposes, with 50% capacity used for research, 25% used by Member States and 25% used for production of operational forecasts.
  • The ECMWF meteorological data archive (MARS) is the largest in the world and continues to grow. As of December 2014, it contains around 68 petabytes of operational and research data, with about 90 terabytes being added daily. More than 180 billion meteorological fields are stored in MARS.

Sign up for our insideHPC Newsletter