Argonne researchers to highlight advances at joint supercomputing/high-performance computing conference in Japan
Researchers from the U.S. Department of Energy’s Argonne National Laboratory will play an active role in the SCA/HPCAsia conference Jan. 26–29 in Osaka, Japan. This event, jointly hosted by SupercomputingAsia (SCA) and The International Conference on High Performance Computing in the Asia-Pacific Region, will bring together experts to discuss advances across applications, artificial intelligence (AI), software sustainability and emerging architectures and technologies. Argonne scientists are participating in a variety of workshops, talks and presentations.
Workshops
Murali Emani has co-organized “WorkshopAI4S,” a full-day workshop on AI and machine learning for scientific applications. Through plenary talks, peer-reviewed papers, keynotes and panel discussions, the workshop will address ways to advance the integration of AI with high-performance computing (HPC) systems and catalyze the next generation of AI-enabled scientific discoveries.
Sheng Di and his colleague from RIKEN have organized a half-day workshop on “Big Data Exploration, Compression and Systems.” The workshop will explore practical approaches to data analytics and mining, big data visualization, data integration and scalable data compression, emphasizing both the characteristics of large-scale data workloads and the constraints of modern hardware architectures. The program includes a paper on checkpoint strategies for large language models (LLMs) by Bogdan Nicolae and Avinash Maurya, along with colleagues from Clemson University, and a paper by Sheng Di and others on compression of pretrained LLMs.
Argonne researchers Justin Wozniak, Nicholas Schwarz and Hannah Parraga have organized “ACX 2026: The First Annual Workshop on Advanced Computing and Experimental Sciences.” This new workshop is intended to provide a forum for researchers and developers of results in “big science” endeavors, with an emphasis on bringing automation into emerging and potentially transformational experimental approaches. Amal Gueroudji will present a paper in this workshop on “Beyond Centralized Labs: Federating the Co-scientist.”
Tutorials
Murali Emani will present a half-day tutorial titled “Programming Novel AI Accelerators for Scientific Computing.” The session will include an overview of AI accelerators at Argonne and hands-on exercises providing attendees with practical experience in refactoring code and running models on these systems.
Anshu Dubey and Akash Dhruv, together with a colleague from Oak Ridge National Laboratory, have organized a tutorial on “Better Software for Science with High-Performance Computing.” Topics range from code design, refactoring and translation to testing of complex scientific software systems. The tutorial will include hands-on activities using LLMs for software development and translations.
Arvind Ramanathan and colleagues from Google Cloud will present a tutorial on “Using AI to Accelerate the Scientific Discovery Pipeline from Deep Research to Development to Data Analysis.” This full-day tutorial will show how AI tools are leveraged at every stage of the research lifecycle — from deep literature search to final data interpretation.
Birds of a Feather Sessions
Argonne researchers and their colleagues will chair three BOFs:
-
Swann Perarnau and Brice Videau — “Energy Efficiency for High Performance Computing and AI Datacenters”
-
Kazutomo Yoshii and colleagues — “Exploring Chip Prototyping as a New HPC Frontier”
-
Murali Emani and Venkat Vishwanath — “Co-designing Future Systems for AI and HPC: Accelerators, Interconnects, and Ecosystems at Scale.”
-
Kazutomo Yoshii and Robert Underwood — “Wafer-Scale Computing for HPC: Current State, Opportunities, and Community Building”
Invited Sessions
Brice Videau will deliver a short presentation as part of an invited session on
“Vision and Strategy: How will supercomputer centers contribute to the future development of HPC/AI?”
Arvind Ramanathan has co-organized an invited panel session on “AI-Powered Pandemic Preparedness: Building a global supercomputing infrastructure for rapid response.” By bringing together computational biologists, HPC center directors, public health officials and AI researchers, this session aims to catalyze international collaboration and establish best practices for using supercomputing as critical infrastructure for global health security. Ian Foster and Rick Stevens will speak about agentic systems for vaccine design.
In an invited co-located event, Charlie Catlett and a colleague from the Institute of Science Tokyo have organized two sessions of the Trillion Parameter Consortium (TPC) — a global initiative that brings together scientists from government laboratories, academia, research institutes and industry to tackle the challenges in harnessing AI for complex scientific and engineering problems. Rick Stevens will join Catlett and others in the opening session. Several Argonne personnel, including Franck Cappello, Anshu Dubey, Robert Underwood, Valerie Taylor and Ian Foster, will discuss work as part of the TPC International Collaborative Initiatives. Anshu Dubey will give an update on a recent workshop on next-generation ecosystems, Arvind Ramanathan will discuss progress in AI for drug discovery, and Laura Schulz and colleagues will speak about the X Quantum Software Stack.
For the full program, see the SCA/HPCAsia website https://www.sca-hpcasia2026.jp/program.html
Argonne National Laboratory seeks solutions to pressing national problems in science and technology by conducting leading-edge basic and applied research in virtually every scientific discipline. Argonne is managed by UChicago Argonne, LLC for the U.S. Department of Energy’s Office of Science.
The U.S. Department of Energy’s Office of Science is the single largest supporter of basic research in the physical sciences in the United States and is working to address some of the most pressing challenges of our time. For more information, visit https://energy.gov/science.