There are no active ads.

Advertisement

U.S. DOE to Spend $425M to Build “Sierra” and “Summit” Supercomputers

by Todd Haselton | November 14, 2014November 14, 2014 7:00 pm PDT

doe-supercomputer

The U.S. Department of Energy (DOE) said Friday it will spend a total of $425 million to build two new super computers and to research advanced supercomputing technologies. The money is being awarded to the Collaboration of Oak Ridge, Argonne, and Lawrence Livermore (CORAL), three U.S. national laboratories.

$325 million will help the department build two computers that will upgrade existing machines at the Oak Ridge National Laboratory (ORNL) and the Lawrence Livermore National Laboratory (LLNL). An additional $100 million will be spent on a new initiative titled “FastForward 2” that will focus on the research of “extreme scale supercomputing technologies.” The DOE did not announce Argonne’s award.

One machine, dubbed Summit, will replace the aging Titan super computer currently used by ORNL, and is expected to offer “at least five times the performance.” Sierra will replace the current Sequoia machine inside LLNL and should be seven times faster, the DOE said. Both computers will feature NVIDIA Volta graphics processors, IBM Power Architecture and connection technology provided by Mellanox.

“High-performance computing is an essential component of the science and technology portfolio required to maintain U.S. competitiveness and ensure our economic and national security,” Secretary Ernest Moniz explained. “DOE and its National Labs have always been at the forefront of HPC and we expect that critical supercomputing investments like CORAL and FastForward 2 will again lead to transformational advancements in basic science, national defense, environmental and energy research that rely on simulations of complex physical systems and analysis of massive amounts of data.”

DOE

Todd Haselton

Todd Haselton has been writing professionally since 2006 during his undergraduate days at Lehigh University. He started out as an intern with...

Advertisement

Advertisement

Advertisement