US funding for supercomputers

17 November 2014

US energy secretary Ernest Moniz has announced major funding packages to support high performance computing projects - $325 million to build new supercomputers at the Department of Energy's (DoE) Oak Ridge and Lawrence Livermore National Laboratories, plus $100 million for R&D into "extreme scale" supercomputing technologies.

The state-of-the-art supercomputers will be five to seven times more powerful when fully deployed than the fastest systems in the US today. They are being built under the joint Collaboration of Oak Ridge, Argonne, and Lawrence Livermore (CORAL) initiative, established earlier this year to leverage supercomputing investments, streamline procurement processes and reduce development costs. Both CORAL awards use IBM Power Architecture, NVIDIA’s Volta GPU and Mellanox’s Interconnected technologies.

Oak Ridge National Laboratory's new system, Summit, is expected to provide at least five times the performance of its current Titan system, installed in 2012.

Summit will be delivered in 2017 under a contract signed with IBM by the DoE's Oak Ridge Leadership Computing Facility (OLCF). According to ORNL, nuclear energy researchers will be able to use the new hybrid CPU/GPU computing system for reactor-scale simulations to allow safe, increased nuclear fuel burn times, power upgrades, and reactor lifetime extensions, and in doing so reduce the volume of used fuel. The system will also support research in combustion science, climate change science and energy storage.

Lawrence Livermore National Laboratory’s new supercomputer, Sierra, is expected to be at least seven times more powerful than the laboratory's current machine, Sequoia. Argonne National Laboratory will announce its CORAL award at a later date, according to DoE.

The second element of Moniz's announcement, known as FastForward 2, seeks to develop critical technologies needed to deliver next-generation capabilities to enable advanced extreme scale computing research and development for the next decade. The joint project between DoE's Office of Science and National Nuclear Security Administration (NNSA) will be led by computing industry leaders AMD, Cray, IBM, Intel and NVIDIA.

High-performance computing (HPC) is essential for carrying out the highly complex applications needed for a range of critical fields, supporting research in areas that rely on simulations of complex physical systems with massive amounts of data including nuclear energy research and environmental monitoring.

Moniz described HPC as an essential component of the USA's science and technology portfolio.

"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 defence, environmental and energy research," he said.

Researched and written
by World Nuclear News