Users of Frontier — the world’s most powerful supercomputer for open science — joined scientific computing experts at the Department of Energy’s Oak Ridge National Laboratory to share recent achievements and new directions in high-performance and quantum computing research at the 2026 Oak Ridge Leadership Computing Facility (OLCF) User Summit, held on June 2-3, 2026.
The two-day event featured user-contributed talks, research posters, scientific data visualization presentations and tours of ORNL facilities, including the new IQM quantum computer installed earlier this year. Much of this year’s program looked ahead, and the discussion focused on the OLCF’s next HPC systems, Lux and Discovery, and how those platforms and the OLCF will support the Genesis Mission — DOE’s effort to connect HPC, quantum computing and AI resources with research facilities across all 17 national laboratories.
“We’re really excited to see so many current, new and prospective users this year,” said Verónica Melesse Vergara, the summit’s lead organizer and head of the OLCF’s HPC Operations section. “We have a vibrant OLCF user community that continues to expand, and this event allows us to learn and showcase how users leverage our capabilities in HPC, AI and quantum computing in their research.”
The 2026 OLCF User Summit welcomed more than 150 scientific computing experts from government, academia, universities and industry. Credit: Alonda Hines/ORNL, U.S. Dept. of Energy
Some 160 participants attended the 22nd annual user meeting, including domestic and international attendees from ORNL, other national labs, universities, industry and federal agencies. The event attracted 61 first-time attendees and 35 prospective users.
HPC here and now
The first day opened with user-contributed talks selected by the OLCF User Group (OUG) Executive Board, who highlighted major scientific achievements enabled by the Frontier supercomputer and OLCF resources. The speakers and their topics included the following:
- Spencer Bryngelson, Georgia Tech – Evaluation and Evolution of AMD Flang Targeting AMD GPU Devices
- Arthur Lorenzon, Universidade Federal do Rio Grande do Sul – Using Omnistat and GPU Power Management to Improve Energy Efficiency on Frontier
- Joe Oefelein, Georgia Tech – Performance Optimization of Turbulent Reacting Flow Simulations on hybrid CPU/GPU Architectures
- Suhas Jain, Georgia Tech – Simulation and AI for Multiphase Turbulence at Extreme Scale Exascale
- Dani Cassol, Lawrence Berkeley National Laboratory – JAWS: Portable Scientific Workflow Execution Across DOE HPC Facilities
The presentations were followed by posters and lightning talks that highlighted work from many early-career professionals as well as a scientific data visualization showcase in the EVEREST visualization facility that featured stunning images and animations and demonstrated the speed and fidelity of the exascale simulations performed on Frontier.
Rohini Vaideswaran presenting her research poster to OLCF User Assistance group leader Chris Fuson (left) during the 2026 OLCF User Summit poster session. Credit: Alonda Hines/ORNL, U.S. Dept. of Energy
The award for best poster went to a team from Georgia Tech. Their poster, Understanding and Analyzing Turbulence Using 32768-Cubed Simulation Data on Frontier, was presented by Rohini Uma-Vaideswaran, Daniel L. Dotson, and P. K. Yeung and was co-authored by Mark Warnecke, P. L. Johnson, C. Meneveau, and S. Nichols. The work represents the first direct numerical simulation of small-scale turbulence captured at an unprecedented resolution of 35 trillion grid points.
Dilip Asthagiri presents Frontier supercomputer simulations on the tobacco mosaic virus during the OLCF data visualization showcase in EVEREST. Credit: Jeremy Rumsey/ORNL, U.S. Dept. of Energy
The award for best data visualization went to ORNL’s Dave Pugmire and Dilip Asthagiri for the Simulation of the Satellite Tobacco Mosaic Virus in Explicit Water. The simulation was originally developed at the University of Illinois Urbana-Champaign around 2006. It represented the first major virus particle simulation when it ran on 256 nodes of the Altix supercomputer. Composed of slightly more than 1 million atoms, the simulation is now routinely run on only a single Frontier node.
Looking ahead to Genesis
On day two, OLCF leadership described the road map for deploying ORNL’s next generation of computing resources, including the leadership-class Discovery supercomputer and the AI + HPC Lux platform. Leadership also focused on how those systems and the OLCF-led American Science Cloud will play key roles in advancing the Genesis Mission as a nationwide effort to integrate HPC, AI, and quantum computing into a cohesive platform for accelerating scientific research.
The OLCF’s Verónica Melesse Vergara (left), Arjun Shankar, and Ashley Barker talk about the future of computing at ORNL and the important role that the OLCF and the American Science Cloud will play in the Genesis Mission. Credit: Alonda Hines/ORNL, U.S. Dept. of Energy
In addition to the day’s events, a closed-session workshop was also held for the OLCF’s Center for Accelerated Applications Readiness program. The workshop focused on the latest developments for the first nine projects that will run on the Discovery supercomputer when it becomes available in 2028.
“The strength of the OLCF lies not only in the systems we operate, but also in the partnerships that help us bring new technologies to the scientific community and explore innovative approaches to advancing research,” said OLCF program director Ashley Barker. “By combining advanced computing, AI, and our staff’s expertise, these collaborations help accelerate discovery in areas such as energy, materials science, manufacturing and other fields critical to the nation’s future.”
User Group elections
Each year at the user meeting, attendees vote to elect new members to serve on the OUG Executive Board for a term of three years. The OUG provides advice and feedback to the OLCF on current operations and future services.
Finishing their terms of service this year were Mike Zingale (Stony Brook University), Emily Belli (General Atomics), and Scott Callaghan (Statewide California Earthquake Center at the University of Southern California). A special thanks was given to outgoing chair Eric Nielsen (NASA).
Reuben Budiardja announces OLCF OUG election results during day two of the User Summit. Credit: Carol Morgan/ORNL, U.S. Dept. of Energy
This year’s new members include Herschel Chawdhry (Florida State University), Daniel Moore (Sandia National Laboratories), and Belli — who was elected to serve an additional term. To learn more about the OUG community and see the current list of board members, visit the OUG website.
What users had to say
Arthur Lorenzon is a professor at the Universidade Federal do Rio Grande do Sul in Porto Alegre, Brazil, and started using Frontier in 2023. Lorenzon’s work focuses on improving the energy efficiency of Frontier and memory-intensive, GPU-accelerated applications.
“We are testing different applications from different research groups, and we are finding that — for example — for some applications, if we can just tune the operating frequency of the GPUs, we can save about 25-percent of the energy without degrading the performance of the applications,” Lorenzon said. “I’d like to thank all the OLCF staff. I’m very happy to collaborate with this group. We are so excited for the new machines, and our goal is to keep collaborating and making better use of the hardware resources so we can reduce energy consumption and make a more sustainable system and a more sustainable facility.”
Eric Nielsen, with Arjun Shankar (left) and Ashley Barker, was awarded a certificate of appreciation for his tenure as Chair of the OUG Executive Board. Credit: Carol Morgan/ORNL, U.S. Dept. of Energy
Senior research scientist Eric Nielsen of NASA’s Langley Research Center in Virginia has also been an OLCF user since 2009 and has been involved in multiple OLCF projects. Among the most recent supercomputing successes, Nielsen highlighted an OLCF project that evaluated retro propulsion for decelerating a human-scale Mars lander concept — a vehicle roughly the size of a two-story house.
“We’ve never flown anything like that, so we really need to determine if we will be able to reliably control the vehicle during descent on Mars in the presence of such a complex and dynamic flow field,” Nielsen explained. “The Summit and Frontier systems ultimately allowed us to start addressing that question after five or six years’ worth of work.”
Nielsen and his team leveraged a closed-loop flight control system running at NASA Langley, which was communicating in real time with Frontier compute nodes, to simulate the high-fidelity, reacting-gas flow physics encountered during a Mars entry trajectory. Nielsen said it was a powerful result that demonstrated the feasibility of developing such a concept at some point in the future.
“But when people ask me what the biggest benefit from the OLCF is, I always first and foremost think of the people and the staff,” Nielsen said. “The OLCF has world-class staff here that you can tap into in so many ways — and not just the support folks and the hardcore computational scientists, but also the administrative folks who set up accounts and manage all the logistics behind the scenes. Everyone is just fantastic across the board.”
The 2026 User Summit planning committee included OLCF staff members Aaron Barlow, Katie Bethea, MacKenzie Boyd, Reuben Budiardja, Dmytro Bykov, Haley Calbaugh, Brian Etz, Chris Fuson, Anjus George, Antigoni Georgiadou, Angela Gosnell, John Holmen, Jazlyn Ilamni, Gina Jaklic, Rajesh Kalyanam, Vanessa Lama, Alan Longcoy, Verónica Melesse Vergara, Suzanne Parete-Koon, Tatiyanna Singleton, Michael Sandoval, and Asim YarKhan.
The OLCF is a DOE Office of Science user facility at ORNL.
UT-Battelle manages ORNL for DOE’s Office of Science, the single largest supporter of basic research in the physical sciences in the United States. The Office of Science is working to address some of the most pressing challenges of our time. For more information, please visit energy.gov/science.
