September 2026 • News
NSDF Team to Lead SC26 Discussion on a Scientific Data Fabric for AI-Driven Discovery
NSDF Team to Lead SC26 Discussion on a Scientific Data Fabric for AI-Driven Discovery
The National Science Data Fabric (NSDF) leadership team will engage the scientific community at SC26 through a Birds of a Feather (BoF) session titled “Building the Scientific Data Fabric Community for AI-Driven Discovery.” The session is part of the ongoing work of the NSF-funded NSDF to connect scientific infrastructure and broaden access to data.
On Sept. 1, 2026, the NSDF transitioned from a pilot project to a production-grade national cyberinfrastructure and service supported by the National Science Foundation with a mission to democratize big data by making it more accessible for scientific discovery. The BoF session will bring together researchers, facility operators, cyberinfrastructure developers, and domain scientists to deepen understanding of that mission and explore the challenges involved in building a more connected scientific data ecosystem.
NSDF Principal Investigator and Co-founder Michela Taufer of the University of Tennessee will lead the session alongside Co-PI and Co-founder Valerio Pascucci of the University of Utah; Co-PI X. Carol Song of Purdue University; Co-PI Virginia Trueheart of the Texas Advanced Computing Center; and David Kanter, Head of MLPerf at MLCommons, whose benchmarking tools are used by the NSDF.
The panel will explore how modern research increasingly requires instruments, data repositories, workflows, AI services, edge devices, HPC systems, and scientists to work together across institutional boundaries. Drawing on examples from astronomy, geosciences, materials science, medicine, and high-energy physics, participants will identify common infrastructure challenges and help define a roadmap for interoperable, federated scientific data ecosystems.
Through the discussion, the NSDF leadership aims to engage the broader scientific community in examining how federated data architectures and autonomous AI could reshape the ways scientific data are accessed, shared, and used for discovery.
Session details: Wednesday, Nov. 18, 2026, 5:15–6:45 p.m. CST, W190b.
Interested in the NSDF’s mission? Consider attending this BoF at SC26:
Autonomous Science Network: A Grassroots Community for Trustworthy Science
This BoF offers a timely opportunity to explore how autonomous experimentation and agentic workflows can accelerate scientific discovery while ensuring new results are robust and verifiable.
Session leader: Rafael Ferreira da Silva.
Additional Session Leaders: Michela Taufer, NSDF PI and Co-founder, Tom Gibbs, NVIDIA, Ravinder Kapoor, Dept. of Energy.
Session details: Wednesday, 18 November 2026, 12:15pm - 1:15pm CST, Location: W183a
Supported by the U.S. National Science Foundation under Award No. 2609465.