World’s fastest supercomputer to be replaced by 5x more powerful machine

The US Department of Energy (DOE) has set its sights on maintaining America’s lead in high-performance computing.

Earlier this month, the agency issued a request for proposals (RFP) for a cutting-edge supercomputer named Discover, touted to replace Frontier, the current fastest supercomputer in the world.

Frontier— housed at Oak Ridge National Laboratory (ORNL) in Tennessee— claimed the top stop on the Top500 list of the world’s fastest supercomputers for the fifth consecutive time in May this year.

The current world champion is built with AMD Epyc processors on HPE Cray EX architecture and sports nearly 8.7 million combined CPU and GPU cores. Additionally, it boasts a high-performance Linpack benchmark score of 1.206 exaflops.

While the DOE has not specified an exact performance target for Discovery, it is expected to deliver three to five times more computational throughput than Frontier, potentially pushing its performance beyond 8.5 exaflops.

“This project is exciting because we will be building something more capable than Frontier, with technologies that will push the edge of what’s possible,” said Matt Sieger, ORNL’s project director for Discovery, in a statement.

A new era of scientific discoveries

Discovery will revolutionize scientific research across various fields and will drive breakthroughs in climate change prediction, drug discovery, high-energy physics, and green energy solutions with its upgrade in computational power.

Speaking about Discovery’s potential, Georgia Tourassi, ORNL’s associate laboratory director of computing and computational sciences highlighted how the scientific community would be able to model real-world situations at new levels of detail.

“It will help us study challenging problems we can’t easily explore with experiment, observation, or theory alone.”

Beyond traditional scientific applications, the supercomputer is designed to excel in advanced artificial intelligence and machine learning tasks, pushing the boundaries of what’s possible in fields like materials science and industrial product design.

Furthermore, Discovery will play a crucial role in the DOE’s Integrated Research Infrastructure (IRI) initiative. This program aims to combine various research tools and scientific facilities.

The road ahead

The DOE has allotted time until August 30, 2024, for interested vendors to submit their proposals for Discovery. The target delivery date to the Oak Ridge Leadership Computing Facility (OLCF) is set for 2027 or early 2028.

Contrary to previous RFPs, the DOE has not specified a speedup goal over Frontier. Instead, the department has outlined a wide range of objectives for the next-generation system. These include improved energy efficiency and full-system modeling and simulation, in addition to advanced AI and machine learning capabilities.

Energy efficiency remains a top priority for the OLCF, which has boosted its computational power 500x while only increasing energy consumption fourfold over the past decade. Once operational, Discovery is expected to be accessible to researchers worldwide, who can compete for computing time to tackle scientific challenges.

Dr. Ceren Susut, associate director for science in the DOE’s Advanced Scientific Computing Research program emphasized the impact of such initiatives. “These discoveries will help shape our understanding of the universe, advance our capabilities for the predictability of the Earth system, bolster US economic competitiveness and contribute to a better future.”