Michela Taufer, MathWorks Professor in the Tickle College of Engineering at the University of Tennessee, Knoxville, will lead a $9 million National Science Foundation project to build a national digital infrastructure that lets researchers securely discover, access, analyze and share scientific data. The National Science Data Fabric (NSDF) aims to remove the data bottlenecks that slow AI-driven discovery across disciplines and institutions.
The bottleneck in data-driven science
Modern scientific instruments generate enormous, heterogeneous data streams. Even researchers with access to high-performance computing can spend months moving and organizing data before analysis begins. Collaborators at smaller institutions often lack the resources to participate at all.
"We are living through a scientific renaissance driven by data and AI," Taufer said. "The challenge is turning all that data into discoveries quickly enough to accelerate scientific innovation." The team includes experts from UT, the University of Utah, Purdue University, the Texas Advanced Computing Center, and MLCommons, with additional partners from national laboratories and industry.
How the National Science Data Fabric works
NSDF connects data where it is generated-a leadership-class computer, a campus cluster, or a specific instrument-directly to researchers, eliminating the need to copy and move massive datasets. A student at a small university could work with petabytes of NASA satellite data, and physicists across the globe could collaborate on shared dark matter datasets in near real time.
Earlier this summer, the team demonstrated this capability. Scientists at the Cornell High Energy Synchrotron Source operated a beamline to examine an additively manufactured stainless steel wall. NSDF linked the experiment to Oak Ridge National Laboratory's AI infrastructure while it ran. ORNL used the streaming data to build and update an AI model of strain inside the metal and sent back recommendations for the next measurement point. Researchers in three states watched the process simultaneously.
"NSDF provides the digital backbone that connects experimental facilities, AI services, computing resources, data repositories and scientists into a unified research ecosystem," Taufer said.
From pilot to production
During its pilot phase, NSDF indexed more than 75 petabytes of data across 68 repositories, showing that secure, cross-institutional sharing is feasible. That volume is a fraction of what instruments produce daily. With the new NSF award, the team will transition NSDF from a successful research prototype to a production-scale national resource serving multiple scientific disciplines.
The shift requires integrating very different facilities-synchrotrons, neutron sources, satellite missions, supercomputers-each with its own data formats, technologies and policies. "A synchrotron, a neutron source, a satellite mission and a supercomputer each generate different kinds of data using different technologies and policies," Taufer said. Building that connected system demands continued advances in cyberinfrastructure, AI and data management, along with a growing community of domain scientists and engineers.
Tennessee at the forefront of AI science
UT has advanced high-performance computing for decades. Through statewide initiatives like AI Tennessee and national efforts like NSDF, the university is shaping AI for Science & Research with partners spanning academia, national labs and industry. The university is also training the next generation in modern cyberinfrastructure and AI-enabled workflows-students and early-career researchers will help design, build and deploy the technologies behind NSDF.
Why this matters for Science & Research
For working scientists and research professionals, NSDF represents a shift in how access to large-scale data and AI compute works. Rather than spending months on data logistics, teams can run experiments and analyses in real time, regardless of their home institution's computing resources. For those building the skills to work in this environment, an AI Learning Path for Research Scientists provides structured training on the AI techniques and workflows that underpin projects like NSDF.
Your membership also unlocks:
