Back to Top
Skip to main content
NETL Logo
NETL-Led Team Nets DOE Award To Advance Cerebras Wafer-Scale Engine Computer Chip for Energy Research Simulations
NETL develops the world’s first computational fluid dynamics simulation on the Wafer Scale Engine, paving the way for a new class of high-performance computing.

NETL develops the world’s first computational fluid dynamics simulation on the Wafer Scale Engine, paving the way for a new class of high-performance computing.

NETL and partner Cerebras Systems of Sunnyvale, California, have been awarded $8 million by the U.S. Department of Energy (DOE) to advance the study of scientific phenomena using the Cerebras Wafer-Scale Engine (WSE).

The third generation WSE-3, which powers Cerebras’ CS-3 AI supercomputer, is an innovation that is designed to enable energy-efficient, extreme-speed scientific simulations, expand NETL’s computational modeling capabilities, reduce its carbon footprint, and evolve DOE’s leadership in traditional high-performance computing and artificial intelligence in a whole new approach.

DOE awarded the funding in response to a DOE laboratory call for proposals for competitive portfolios to advance scientific computing research.

According to Tammie Borders, the NETL project lead, the Cerebras WSE-3 represents a miniaturized and optimized supercomputer on a single giant silicon wafer. She said the project has three specific objectives:

  • Demonstrate energy-efficient, extreme-speed scientific simulation capability for distributed memory problems.
  • Implement computer science methods to extend the WSE-3 to scientific computing and grid challenges.
  • Develop methods to extend high-bandwidth, low-latency communications.

The project builds upon ongoing efforts by NETL and Cerebras to use the WSE-3 to accelerate innovation across industrial and energy systems from fuel production and combustion to thermal management.

Early-stage results from the five-year NETL Cerebras WSE-3 research partnership demonstrated promising results for a new path to exascale computing and beyond. According to DOE, exascale computers are digital computers, like today’s computers and supercomputers but with much more powerful hardware.

“Conventional approaches requiring data movement have become a significant performance bottleneck,” Borders explained. “The NETL-Cerebras approach is expected to realize a concept called ‘near-memory computing,’ that incorporates memory and logic in an advanced hardware package.”

NETL researcher Dirk Van Essendelft, who spearheaded the novel research activities, said NETL will extend its existing research from a single WSE to WSE clusters while maintaining performance and energy efficiency gains.

“It all means that our team is developing a transformative scientific computing approach that can tackle complex energy and security challenges that are well beyond what is capable of the most advanced computer today,” he said.

He said that he believes the WSE-3 can achieve significant speed gains by system architecture miniaturization.

Borders added that, “The WSE-3 integrates memory and processing on the same chip, eliminating wasted compute cycles and lowering net data access energy.”

Currently, research computing represents about half of NETL’s total energy consumption and CO2 emissions, which are representative of data center and computing growth worldwide.

“In this new work, we believe NETL and Cerebras are developing a transformative scientific computing approach to tackle complex energy and security challenges that is well beyond what is capable of the most advanced computer today.” said van Essendelft. 

Cerebras Systems designed its WSE hardware to tackle tough artificial intelligence problems. Several years ago, an NETL research team realized that this new type of computer chip could be used to solve real-world engineering problems and engaged with the company to use its approach for NETL energy research.

“Our multi-year partnership with NETL has yielded incredible scientific breakthroughs across computational fluid dynamics and the modeling of energy systems – outperforming the Joule 2.0 supercomputer by 470x -- while significantly decreasing NETL’s total energy consumption and CO2 emissions,” said Andy Hock, SVP of Product and Strategy, Cerebras. “We look forward to applying the power of our wafer-scale AI technology to tackle scientific computing and grid challenges with this new DOE grant.”

Cerebras is a private, U.S.-owned computer chip company founded in 2016 that has a global footprint.

NETL is a DOE national laboratory that drives innovation and delivers solutions for an environmentally sustainable and prosperous energy future. By using its world-class talent and research facilities, NETL is ensuring affordable, abundant, and reliable energy that drives a robust economy and national security, while developing technologies to manage carbon across the full life cycle, enabling environmental sustainability for all Americans.