NSF Leadership-Class Computing Facility
Discovery at the frontiers of science and engineering requires computing power at a massive scale, sustained over a long period of time. To reach this goal, the U.S. National Science Foundation (NSF) is funding the design of a Leadership-Class Computing Facility (LCCF) through its Major Research Equipment and Facilities Construction (MREFC) process, with an operations model on a multi-decade scale.
Computation – through large scale simulation, data analysis, and artificial intelligence applications – is essential across many areas of research and development. The LCCF will enable transformative discoveries for broad classes of curiosity-driven and use-inspired science and engineering applications.
The LCCF is expected to begin operations during 2026 and will deploy the largest academic supercomputer dedicated to open-scientific research in the NSF portfolio. The supercomputer, called Horizon, will provide 10x performance improvement for simulation over the current NSF Leadership-Class Computing system, Frontera, as well as meet the unique scientific requirements of the NSF community.
For AI applications, the leap forward will be even larger, with more than 100x improvement over Frontera. Horizon will include a significant investment in specialized accelerators to enable state-of-the-art artificial intelligence research, as well as more general-purpose processors to support the diverse needs for simulation-based inquiry across all scientific disciplines.
The LCCF is also envisioned as a key element in advancing a future National Artificial Intelligence Research Resource (NAIRR), which is currently being piloted by NSF in coordination with other federal agencies and non-governmental partners. The NAIRR is envisioned as a widely accessible, national cyberinfrastructure that will advance and accelerate the U.S. AI R&D environment and fuel AI discovery and innovation in the United States.
Key Features of the LCCF
Large-scale data storage systems
Wide suite of software and services
Four distributed science centers across the US
Wide range of education and outreach plans
Learn More
The LCCF has been designed through an intensive process striving to reflect the collective requirements and insights of the community. Multiple workshops were held to determine what science researchers hope to accomplish over the next 10 years, and what technologies users believe are critical to the success of the design. These reports summarize the community response.
Report of the Workshop in Future Directions in Extreme Scale Computing for Scientific Grand Challenges
Held 09-10 January 2020 Texas Advanced Computing Center The University of Texas at Austin
Omar Ghattas, George Biros, Dan Stanzione, Rick Stevens, John West
Summary of Community Inputs from SC19
A Report from the LCCF Birds of a Feather Session Held 21 November 2019, Denver CO
Omar Ghattas, Dan Stanzione, Rick Stevens, John West
Contact Us
For more information, contact lccf-community@tacc.utexas.edu.
We Need Your Input!
Designing the LCCF is an intensive process that will only be successful if the design reflects the collective requirements and insights of the community. What science will you need to accomplish over the next 10 years? What technologies do you think are critical to the success of the design?
Please consider adding your voice to the process by attending a future LCCF planning event, or by sending your thoughts to lccf-community@tacc.utexas.edu.
Past Workshops
January 9-10, 2020
Texas Advanced Computing Center
1/9 full day; 1/10 half day
Austin, TX
Science Requirements Workshop
November 21, 2019
SC19 Birds of a Feather
Denver, CO
Science and Technology Requirements for Leadership Computing in Open Science
For information on attending future workshops, please send an email to lccf-community@tacc.utexas.edu.
A description of the science requirements driving the development of the facility
A description of the research infrastructure and technical requirements needed to meet the science objectives
A system-level design, including definition of all functional requirements and major subsystems
Identification and evaluation of enabling technologies for scientific productivity of the next generation leadership-class system
A description of proposed education and industry outreach
Broader societal impact activities for the future facility