The Department of Energy’s (DOE) office of science and the national
nuclear security administration (NNSA) have awarded $25.4 million in
research and development contracts to five leading companies in
high-performance computing (HPC) to accelerate the development of
next-generation supercomputers.
Under DOE’s new DesignForward initiative, AMD, Cray, IBM, Intel
Federal and Nvidia will work to advance extreme-scale, on the path to
exascale, computing technology that is vital to national security,
scientific research, energy security and the nation's economic
competitiveness.
Exascale Computing Is Crucial
“Exascale computing is key to NNSA’s capability of ensuring the
safety and security of our nuclear stockpile without returning to
underground testing. The resulting simulation capabilities will also
serve as valuable tools to address nonproliferation and counterterrorism
issues, as well as informing other national security decisions,” said
Robert Meisner, director of the NNSA office of advanced simulation and
computing program.
As the nation’s largest funder of physical science research, DOE
provides thousands of researchers at national labs and universities with
access to some of the world’s most powerful supercomputers. These
systems, which have peak speeds of quadrillions of calculations per
second, are helping scientists study climate change, develop renewable
energy sources, understand the makeup of our universe and develop new
materials. But taking these research missions to the next level will
require supercomputing systems that are 1000 times faster than today’s
systems.
“In an era of fierce international HPC competition, the development of
exascale computing becomes critical not only to our national security
missions but to the nation’s economic competitiveness in the global
marketplace. This partnership between industry, the DOE Office of
Science and NNSA supports the development of technology to overcome the
obstacles on the road to exascale systems,” said William Harrod,
FastForward program manager and research division director for DOE’s
advanced scientific computing research program.
DesignForward Initiative
The DesignForward contracts, which cover a two-year performance
period, will support the design and evaluation of interconnect
architectures for future advanced HPC architectures. Such interconnects
will tie together hundreds of thousands or millions of processors, as
building blocks of supercomputers to be used in studying complex
problems in unprecedented detail. The DesignForward focus will be on
developing interconnects that are energy efficient, have high bandwidth,
and minimize the time to move data among processors.
“A major disruption is facing high performance computing because
energy constraints are causing our building blocks, microprocessors and
memory to change dramatically. We need to collaborate with computer
companies to ensure that future supercomputers meet DOE's mission needs
in science, energy and national security. Berkeley Lab is pleased to
place these contracts on behalf of DOE and its laboratories,” said Sudip
Dosanjh, director of the national energy research scientific computing
center at Lawrence Berkeley national laboratory (Berkeley Lab).
DesignForward is the follow-on to DOE’s FastForward project, a
public-private partnership between DOE and HPC industry to advance
extreme scale computing technologies with the ultimate goal of funding
innovative, critical R&D technologies needed to deliver next
generation capabilities within a reasonable energy footprint.
FastForward is funded by DOE's Office of Science and NNSA, and
technically managed by seven national laboratories.
Successful Public-Private Partnership
Under the new contract, Intel will focus on interconnect
architectures and implementation approaches, Cray on open network
protocol standards, AMD on interconnect architectures and associated
execution models, IBM on energy-efficient interconnect architectures and
messaging models and Nvidia on interconnect architectures for massively
threaded processors.
“We are honored to be selected for this research program and to
continue our work with DOE to shape the future of high-performance
computing, started last year with FastForward. This public-private
partnership extends AMD’s research into the use of next generation APUs
to meet the demands of extreme-scale computing. We believe this program
will benefit the DOE and provide technology insights into challenges
throughout the computing industry,” said Alan Lee, AMD’s corporate vice
president of research and advanced development.
The vendors will collaborate with DOE's exascale co-design centers to
determine how changes in the system architectures will affect how well
the scientific applications perform.
“U.S. leadership in HPC is essential to meeting the mission-critical
needs of DOE and other federal agencies. We are proud to have been
chosen by DOE as a continued partner in their strategic work toward the
advancement of next-generation supercomputing technology. Intel has a
long standing commitment to science, research and innovation, from
supercomputing to personal computing,” noted Dave Patterson, president
of Intel Federal.
“Partnerships and collaborations are an important element of
exploring new ideas and overcoming the challenges of exascale computing,
and we look forward to working with the DOE researchers and playing a
role in the co-design efforts that will be key to the success of the
DesignForward program,” said Peg Williams, Cray’s senior vice president
of high performance computing systems.
“Exascale computing is vitally important for U.S. scientific research
and economic competitiveness, and Nvidia’s expertise in
massively-parallel heterogeneous systems will play a critical role in
reaching exascale. Interconnection networks are a key component of
exascale systems. We are excited to collaborate with other DesignForward
recipients to enable an open ecosystem for next-generation
high-performance system interconnects,” noted Bill Dally, chief
scientist and senior vice president of research at Nvidia
By: Anton Shilov
Link:
http://www.xbitlabs.com/news/other/display/20131115202502_AMD_Cray_IBM_Intel_and_Nvidia_Receive_25_4_Million_in_Contracts_for_Exascale_Supercomputer_Interconnect_Design.html
