US Department of Energy and AMD Collaborate on New Supercomputers

The US Department of Energy has announced its collaboration with AMD to create two new supercomputers at Oak Ridge National Laboratory. This initiative is part of a broader strategy to enhance research in science, energy, and national security. These two systems represent a public-private investment of approximately one billion dollars, aimed at supporting artificial intelligence research using a modular infrastructure designed and built in the United States.

The Importance of Supercomputers in Scientific Research

The development of supercomputers is a pivotal step in advancing scientific research, as these systems provide immense computational capabilities that enable the processing of vast amounts of data quickly and efficiently. This enhances scientists’ ability to conduct complex simulations that might be impossible under normal circumstances.

Supercomputers are vital tools in various scientific fields such as physics, chemistry, and biology. They allow researchers to analyze scientific data faster, leading to accurate and reliable results in less time.

Lux AI Project: America’s First AI Factory

The Lux AI project is expected to begin operations in early 2026, marking the establishment of the country’s first “AI factory.” This system aims to train and deploy advanced AI models in the fields of science, energy, and security. It is being developed in collaboration with AMD, Oracle Cloud Infrastructure, and Hewlett Packard Enterprise.

Lux AI utilizes advanced technologies like AMD Instinct MI355X processors, EPYC processors, and Pensando networks to handle heavy AI tasks. The architecture is designed to allow multiple groups to work together while maintaining data security and separation.

Discovery Project: Enhancing Computing and AI Strategy

The Discovery supercomputer is expected to be ready by 2028 and will become the main computer for the Department of Energy at Oak Ridge Laboratory. Discovery will use the new EPYC “Venice” series processors and MI430X graphics processors designed for scientific computing and AI.

Discovery features a “bandwidth everywhere” design that enhances memory and network performance without increasing power consumption, allowing it to process more data and run complex models efficiently.

Partnerships and Their Role in Innovation

Companies like AMD, HPE, and Oracle play key roles in building and supporting these systems. Leaders from these companies have stated that the collaboration will help achieve unprecedented productivity and provide high-performance AI infrastructure.

These partnerships serve as a model for private institutions, demonstrating how expertise can be integrated across sectors to develop shared infrastructure while maintaining data control.

Conclusion

The Lux and Discovery projects are significant steps towards enhancing supercomputing and AI capabilities in the United States. They contribute to accelerating research and achieving innovation in diverse fields such as energy, medicine, material sciences, and national defense. These projects reflect how a future AI strategy can transform infrastructure into a long-term competitive advantage.