NVIDIA's AI Microservices and Holoscan Platform Supercharge Materials Science at SC25

Scientists chasing breakthroughs in liquid-cooled data centers, high-resolution displays, and long-lasting batteries need novel chemicals and materials optimized for energy efficiency, durability, and performance. At the SC25 supercomputing conference in St. Louis, NVIDIA unveiled accelerated data processing pipelines and AI microservices designed to supercharge this research, with applications spanning aerospace, energy, and manufacturing.

A standout demo in the NVIDIA booth highlighted work from the U.S. Department of Energy’s Brookhaven National Laboratory, leveraging the NVIDIA Holoscan AI sensor processing platform to visualize materials at under 10 nanometer resolution. Another demo showcased upcoming NVIDIA NIM microservices for high-throughput simulations in conformer search and molecular dynamics—key processes for predicting atomic-level material properties. These microservices form part of NVIDIA ALCHEMI, a suite of tools tailored for chemistry and materials science.

Early adopters include Japanese energy giant ENEOS and New Jersey's Universal Display Corporation (UDC), both pushing the boundaries of energy conversion and OLED technology.

Brookhaven National Laboratory's Real-Time Nanoscale Imaging

Brookhaven's National Synchrotron Light Source II (NSLS-II) uses powerful X-rays to image complex materials like batteries, microelectronics, and nanoparticle superlattices at nanometer scales. These experiments flood researchers with massive datasets requiring advanced computation to yield insights.

By integrating NVIDIA Holoscan, NSLS-II achieves high-bandwidth, real-time edge processing of streaming data. This delivers near-instant feedback during scans, enabling on-the-fly identification of regions of interest and observation of property evolution—crucial for experimental decision-making.

“By collaborating with NVIDIA to integrate Holoscan into our optimize operating costs of expensive instruments like NSLS-II.

“If we can run our experiments more efficiently, we can support more users, which in turn means we can do more science,” said Daniel Allan, group leader of data engineering at NSLS-II. “We also see the potential to use this pipeline for AI-assisted operation — integrating AI models for both imaging tasks and controls to conduct autonomous experiments.”

ENEOS Accelerates Energy Innovation

ENEOS is harnessing ALCHEMI NIM microservices for discovering immersion cooling liquids for next-gen data centers and catalysts for hydrogen fuel production. These tools enable prescreening of molecular candidates via computational experiments, drastically reducing real-world testing needs and R&D costs.

With ALCHEMI, ENEOS evaluated 10 million liquid-immersion candidates and 100 million oxygen evolution reaction candidates in weeks—10x more than previous methods.

“We hadn’t considered running searches at the 10-100 million scale before, but NVIDIA ALCHEMI made it surprisingly easy to sample extensively and achieve more physically realistic results,” said Takeshi Ibuka, general manager of the AI innovation department at ENEOS Holdings, Inc.

Universal Display Corporation Redefines OLED Discovery

UDC, a leader in energy-efficient OLED materials for devices from smartwatches to VR headsets, faces a vast chemical space—around 10^100 possible molecules. ALCHEMI's AI-accelerated conformer search microservice lets UDC evaluate billions of candidates up to 10,000x faster than traditional CPU methods.

Promising compounds then undergo molecular dynamics simulations, accelerated up to 10x per simulation and further scaled across multiple GPUs, slashing times from days to seconds. UDC targets blue phosphorescent OLEDs to boost energy efficiency and performance.

“The NVIDIA ALCHEMI microservices enable more creativity for individual scientists by removing any concerns that we have about capacity and throughput limitations," said Julie Brown, executive vice president and chief technical officer at UDC. “Through this collaboration with NVIDIA, we can amplify the impact of our scientific insight and significantly increase the pace at which new materials are discovered and developed. These efforts don’t just push the boundaries of what OLED can do — they set the stage for more sustainable and energy-efficient displays worldwide.”

NVIDIA ALCHEMI joins over 150 CUDA-X libraries accelerating science and engineering challenges. These advancements at SC25 underscore how GPU-accelerated AI is not merely speeding up computations but reshaping how researchers tackle the molecular frontier, paving the way for technologies that demand ever-smaller, ever-efficient materials.

Source: NVIDIA Blog