News Release

DOE, ORNL announce opportunity to define the future of high-performance computing

R&D projects will engage vendors for forward-looking technologies and energy efficiencies

Grant and Award Announcement

DOE/Oak Ridge National Laboratory

Defining the future of HPC

image: 

The R&D projects selected will engage vendors for forward-looking technologies and energy efficiencies.

view more 

Credit: ORNL

The U.S. Department of Energy’s (DOE) Office of Science (SC) today announced a new research and development opportunity led by Oak Ridge National Laboratory (ORNL) to advance technologies and drive new capabilities for future supercomputers. This industry research program worth $23 million, called New Frontiers, will initiate partnerships with multiple companies to accelerate the R&D of critical technologies with renewed emphasis on energy efficiency for the next generation of post-exascale computing in the 2029 and beyond time frame.

“There is a growing consensus that urgent action is needed to address an array of bottlenecks in advanced computing, including energy efficiency, advanced memory, interconnects and programmability to maintain economic leadership and national security” said Ceren Susut, Associate Director of the Energy Department’s Office of Science for Advanced Scientific Computing Research. “Through the Exascale Computing Project, we partnered with industry to provide critical advances for the Nation, but much more work remains to be done and the opportunities and challenges of AI underscore the need for continued government investment.”

High-performance computing (HPC) and data-driven modeling and simulation are used extensively to advance DOE science missions and are key investment areas for the Office of Science. In fact, SC is home to three of the top ten fastest supercomputers in the world, all used for open science research, including the world’s first and second fastest systems according to the June 2024 TOP500 list of the world’s most powerful supercomputers. These systems include the 1.206 exaflops HPE Cray EX Frontier at ORNL; the 1.012 exaflops HPE Cray EX- Intel Aurora at Argonne National Laboratory; and the 148.6 petaflops IBM AC922 Summit at ORNL.

But the HPC landscape is changing quickly and evolving. New Frontiers provides an opportunity to address and rapidly adapt to those changes.

“With Dennard scaling long dead and the slowing of Moore’s law, we’re seeing technologies critical to HPC consuming more power that partially offset increases in application performance due to improvements in silicon process nodes and improved packaging techniques,” said Christopher Zimmer, New Frontiers project director. “Current technology trends threaten to have a disruptive and costly impact on the development of DOE applications and potentially a negative impact on the productivity of DOE scientists.”

For technologies targeted for productization in the 5 to 10-year timeframe, industry leaders have noted that processors cannot get faster without producing more heat (Dennard scaling) and computer chips likely cannot get much smaller (Moore’s Law). The New Frontiers project’s strategic R&D investment and innovative vendor engagement opportunities will accelerate development of technologies needed for the next generation of exascale computing within these new realities.

"New Frontiers will be exploring technologies to improve energy efficient computing," said Al Geist, director of the Frontier project. "Energy efficiency is becoming critical to building future generations of Leadership computers and will involve development of new approaches to hardware, software, and application algorithms."

New Frontiers will push strategic advancements in HPC in the exascale era, which will be essential in helping DOE maintain leadership and to address the future challenges in science, energy, health, and growing security threats. The science potential of emerging computing systems and other novel computing architectures will require numerous significant modifications to today’s tools and techniques to deliver on the promise of science.

Through this RFP, the ORNL-led team is soliciting innovative R&D proposals in the areas of hardware technologies, software technologies, and cross-cutting technologies.

Due to the emphasis on extreme-scale application workflows, overall time to solution is also an important consideration. The focus of this program will be on open-source and sustainable software technologies for extreme scale HPC systems and the development of techniques necessary to support emerging workloads of integrated facilities across the DOE landscape.

“New Frontiers comes at a critical moment in high-performance computing,” said Ashley Barker, OLCF program director and Exascale Computing Project deputy director. “We have an exciting opportunity to take advantage of the incredible portfolio of applications and technologies created and enabled by the Exascale Computing Project. With New Frontiers we want to translate those advancements to benefit next-generation systems.”

New Frontiers seeks to fund new and/or accelerated R&D of technologies targeted for productization in the 5 to 10-year timeframe and includes a 40% cost share option. The period of performance for any subcontract resulting from this RFP will be two years.

The RFP is being managed by UT-Battelle LLC, which operates Oak Ridge National Laboratory on behalf of the Department of Energy.

Funding for the RFP is being provided by the DOE Office of Science’s Advanced Scientific Computing Research Program.

The full New Frontiers RFP can be found https://www.olcf.ornl.gov/newfrontiers/.

UT-Battelle manages ORNL for DOE’s Office of Science, the single largest supporter of basic research in the physical sciences in the United States. DOE’s Office of Science is working to address some of the most pressing challenges of our time. For more information, visit energy.gov/science.


Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.