NERSC Breaks Petaflop Barrier With ‘Hopper’ Supercomputer
The Department of Energy’s National Energy Research Scientific Computing Center (NERSC) announced on Tuesday that it is now home to the fifth most powerful supercomputer in the world.
The new supercomputer can perform 1.05 petaflops (quadrillions of calculations per second) running the Linpack benchmark.
The machine is named Hopper after Admiral Grace Hopper, a pioneer in software development and programming languages.
The Hopper is funded by Department of Energy’s (DOE) Office of Advanced Scientific Computing Research and is the second fastest supercomputer in the U.S.
NERSC was established in 1974 and provides computing systems and services to over 3,000 researchers supported by the DOE. Its users have reported to produce over 1,500 scientific publications each year as a result of calculations run at NERSC.
“While we are elated to have entered the petascale performance arena, we are especially excited by the computational science potential offered by Hopper,” Kathy Yelick, Director of the NERSC Division and Associate Laboratory Director of Computing Sciences at Berkeley Lab, said in a statement.
“We selected Cray as the system vendor after a competitive procurement based in large part on how proposed systems performed running our application benchmarks. Now that the system is installed and operational, we will begin our acceptance testing in which we run some of the most demanding scientific applications to ensure that Hopper will meet the day-to-day demands of our users.”
NERSC serves one of the largest research communities in the U.S. The center’s supercomputers are used to calculate a wide range of scientific challenges such as global climate change, combustion, clean energy, new materials, astrophysics, genomics, particle physics and chemistry.
Supercomputers are increasingly being used to help out scientific experiments by allowing researchers to test theories using computational models and analyze large scientific data sets.
NERSC also houses Franklin, a 38,128 core Cray XT4 supercomputer with a Linpack performance of 266 teraflops. Franklin ranks as number 27 on the TOP500 list of the world’s fastest supercomputers.
On the Net: