Abstract

"Digital twins are revolutionizing industry, transforming science, global policy, climate change and healthcare. This half day workshop first surveys the state of the art in digital twins then dives into a focused look at digital twins of HPC data centers. Evolving from basic 3D models to highly accurate replicas, digital twins serve as reliable data sources for decision-making, ranging from subatomic to cosmic scales. Applications in computational biomedicine, energy, and automation hint at future roles in quantum computing. The first set of talks seeks to foster a digital twins community in HPC, highlighting the potential to drive breakthroughs and inform critical policy decisions. Pivoting to HPC infrastructure, digital twins are transforming data centers, providing precise virtual models for enhanced efficiency. These models enable exploration of "what-if" analyses and complex behaviors across disciplines. The second set of talks and panel  addresses power and cooling efficiency, network behavior, and AR/VR applications, aiming to advance energy-efficient, sustainable computing solutions for next-generation infrastructures. The concluding panel will address all of the areas touched on and will also include discussion of the role of AI in the future of Digital Twins. 

National Academy of Sciences Engineering & Medicine - definition of Digital Twin (source doc here, sub-committee reports here)

"A digital twin is a set of virtual information constructs that mimics the structure, context, and behavior of a natural, engineered, or social system (or system-of-systems), is dynamically updated with data from its physical twin, has a predictive capability, and informs decisions that realize value. The bidirectional interaction between the virtual and the physical is central to the digital twin"

Link to official ISC26 Workshop Schedule and description

Workshop Agenda & Schedule

2:00 PM - 2:10 PM - Barton Fiske, NVIDIA -> Introductory remarks & Digital Twin State of the Art - Dr. Michael Grieves, UCF 

2:10 PM - 2:35 PM - Dr. Abani Patra, Tufts University -> Neural operators and UQ in Building Digital Twins for Offshore Wind Turbines 

2:35 PM - 3:00 PM - Francois Mazen, Director SciViz, Kitware -> Scientific Visualization Tools for Digital Twins

3:00 PM - 3:25 PM - Dr. Peter Coveney, University College London -> Virtual Humans & Digital Twins in Healthcare

3:25 PM - 3:50 PM - Dr. Dimitrios Rovas, University College London -> The Building of the Future: Digital Building Twins and AI Agents for Smart Operations 

3:50 PM - 4:00 PM - Open QA - All presenters

4:00 PM - 4:30 PM - Break

4:30 PM - 5:00 PM - Dr. Anuj Kapadia, ORNL -> HPC-focused challenges in developing and deploying a human digital twin ecosystem 

5:00 PM - 5:30 PM - Dr. Andrea Townsend-Nicholson, University College London  -> Personalised Healthcare Transformation Through New Tech & Next-Gen Deep Learning

5:30 PM - 6:00 PM  Closing Panel - all presenters 

6:00 PM End of Workshop 

Workshop Presenters:

Dr. Abani Patra

In this talk we will discuss approaches to building digital twins of large offshore wind turbines under complex loading regimes using a combination of HPC based simulations ensembles, data from the turbines and new modeling and UQ approaches.

Bio: Abani Patra researches computational issues in large scale and data driven modeling. He is trained in computational mathematics and has made fundamental contributions to several areas related to modeling complex physical systems including adaptive meshing, HPC, uncertainty quantification and the cyberinfrastucture needed for using such models. His more recent work is focused on AI driven modeling of such systems. He was the founding director of the Institute for Computational and Data Sciences at University at Buffalo and the Data Intensive Studies Center at Tufts University and has directed programs at the National Science Foundation and the U.S. Department of Energy.

Dr. Francois Mazen

Bio: François Mazen is the Director for the Scientific Visualization team at Kitware Europe - 

Throughout his career, François has focused on developing software solutions for scientific communities from research to industry. This topic includes numerical simulation methods, performance, Artificial Intelligence and large software architectures. In 2008, François received his engineering degree at IFMA (French Institute for Advanced Mechanics) in Clermont-Ferrand where he was nominated for TOP 10 students. The same year, he also received a Master of Science at Université Blaise Pascal (Clermont-Ferrand) where he specialized in Rigid Body. Before joining Kitware, his previous 13 years of experience included 4 years at Ansys as a Software Developer to customize the Ansys Workbench application for specific customer requirements. He also worked for more than 6 years at Siemens PLM as Project Leader and Software Architect, where he integrated Robot’s Path Planning Technology in large applications like Siemens NX, Dassault Catia and Tecnomatix Process Simulate. François is an open source enthusiast as he has been contributing to several open source softwares since 2007. Currently, he is an active member of the Debian Operating System development team. Since 2021 he has led Kitware’s European Scientific Visualization team, including team management, strategy, technical expertise, operational excellence and business development.

Professor Peter Coveney

Prof Peter V. Coveney holds a chair in Physical Chemistry, is an Honorary Professor in Computer Science at University College London (UCL), a Professor in Applied High Performance Computing at the University of Amsterdam (UvA), and Professor Adjunct at Yale University School of Medicine (USA). He is Director of the Centre for Computational Science (CCS) at UCL. Coveney is active in a broad area of interdisciplinary research including condensed matter physics and chemistry, materials science, as well as life and medical sciences in all of which high performance computing plays a major role. He has published more than 400 scientific papers and co-authored two best-selling books (The Arrow of Time and Frontiers of Complexity, both with Roger Highfield) and is lead author of the first textbook on Computational Biomedicine (Oxford University Press, 2014). Coveney is a founding member of the UK Government’s E-Infrastructure Leadership Council and a Medical Academy Nominated Expert to the UK Prime Minister's Council for Science and Technology on Data, Algorithms and Modelling which has led to the creation of the London based Turing Institute. He is also a member of Academia Europaea, as well as the London Centre for the Theory and Simulation of Materials, The Thomas Young Centre. 

Dr. Anuj Kapadia

Anuj J. Kapadia is a Distinguished Research Scientist and Section Head for Advanced Computing in Health Sciences at Oak Ridge National Lab (ORNL). He is also an Adjunct Professor of Radiology, Physics, and Medical Physics at Duke University. He holds a Ph.D. in Biomedical Engineering from Duke University. His work focuses on AI, ML, and simulation and modeling in health applications. Dr. Kapadia has over 20 years of experience in neutron and X-ray scattering, Monte Carlo simulation development, and data analytics for security and medical applications. His work has been funded by the DOE, DOD, DHS, NIH, and the NC Biotech Center. His work has received acclaim in press articles, national and international conferences and he has won awards for his teaching, mentorship, and leadership roles. He has mentored 33 students and post-docs and has authored over 100 journal and conference papers. Dr. Kapadia is a member of the Institute for Electrical and Electronics Engineers, the Society for Photo-optical Instrumentation Engineers, and the American Association for Physicists in Medicine professional societies.

Dr. Dimitrios Rovas

Dimitrios Rovas is Professor of Building Simulation and Optimisation at the Bartlett School of Environment, Energy and Resources, University College London, where he developed and leads the MSc in Smart Buildings and Digital Engineering and heads the corresponding research team. He has spent the last fifteen years researching smart buildings, with a focus on computational approaches to energy modelling, semantic data integration, and the middleware that turns building data into operational intelligence. He received his PhD in Mechanical Engineering from MIT and previously held academic posts at the University of Paris VI, the University of Illinois at Urbana-Champaign, and the Technical University of Crete. He has led more than twenty major research projects advancing data-driven platforms for building operation, and co-led the International Energy Agency's Annex 81 on Data-Driven Smart Buildings, a collaboration spanning nineteen countries. He is co-founder and chairman of Tacit Analytics, a UCL spinout developing reusable AI agents for smart buildings and building decision-making, where research on semantic digital twins is being translated into commercial

practice.

Workshop Organizers:

Andrea Townsend-Nicholson

Bio: Andrea holds a chair in Biochemistry & Molecular Biology at University College London (UCL). She obtained a BSc degree in Molecular Genetics & Molecular Biology (University of Toronto, Canada; 1986) and a DSc degree in Cellular & Molecular Biology (Université Louis Pasteur, France; 1990), completed postdoctoral training at the Garvan Institute of Medical Research (Sydney Australia;1991-1995) and UCL (1995-1999), and held a British Heart Foundation Fellowship at UCL from 1999-2001. Her research focuses on understanding the molecular basis of cell surface receptor function in health and disease using a combination of experimental and computational methodologies. Andrea is particularly interested in facilitating the introduction of personalised medicine into clinical practice and in the development of computational methodologies that converge with experimental findings. She teaches medical and undergraduate bioscience students to use supercomputers as part of their taught university curriculum.

Andrea's presentation - Digital Twins – Enabling Personalised Healthcare Transformation Through New Technology and Next-Generation Deep Learning

We currently lack an R&D framework that delivers an in silico-to-in vitro-to-in vivo continuum for truly effective healthcare solutions – those that personalise care, identify and prevent adverse drug reactions, and optimise treatment strategies. Such a framework requires unconventional approaches capable of capturing dynamic processes across preclinical and clinical domains to bridge the gap from personalised screening to clinical insights. By using adaptive deep learning and integrated screening technologies for continuous dataflow and real-time predictions, digital twins can act as integration hubs linking in silico and in vitro with extensions to in vivo via closed loop operations, allowing individual R&D processes to be controlled and extending diagnostic capabilities into patient care. 

Barton Fiske

Bio: As senior alliances and product manager for math libraries, devtools and digital twins at NVIDIA, Barton has been fascinated by 3D graphics, visualization and computer gaming from a very early age and pursued his degree in Computer Science specifically to further these interests into a full blown profession. A graduate of the Rochester Institute of Technology, Barton has more than 30 years experience in a variety of prior roles, ranging from software engineer to systems engineer and demo architect to senior cloud architect and director of technical sales and evangelism. Barton is the co-author of two books on Java programming and has developed dozens of interactive 2D, 3D and VR demo experiences for a wide variety of industrial scenarios, from global product launches to end user applications. He is currently using Digital Twin technology in his own life to manage and maintain models of his seaside home in Newport, RI. 

Workshop steering comittee / advisors : (will not be in person @ workshop)

Peter Messmer - Director of Developer Technology - Digital Twins, HPC Viz, NVIDIA - co-chair

Bio: Peter is a senior manager in the HPC Developer Technology group at NVIDIA. He and his team work on tools and technologies to help clients use GPUs to accelerate their scientific discovery processes. Peter holds an MSc and PhD in physics from ETH Zurich, Switzerland, with a specialization in kinetic plasma physics and nonlinear optics.

Wes Brewer - Senior Researcher AI for Science & Digital Twins, Oak Ridge National Laboratory - co-chair

Bio: Wesley Brewer is a Senior Research Scientist in the Analytics & AI Methods at Scale (AAIMS) group at Oak Ridge National Laboratory. He holds a PhD in Computational Engineering from Mississippi State University and a MS in Ocean Engineering from MIT. Wesley's background is in computational fluid dynamics (CFD), he has considerable experience in modeling and simulation, as well as machine learning for science and engineering applications at scale on HPC. He has developed several CFD codes the U.S. Navy utilized in the design and analysis of advanced propulsion systems. Moreover, he’s developed a number of machine-learned models to support the U.S. Army’s helicopter program as well as the DoD in general, for which he won the 2019 DoD HPCMP Hero Award. Recently joining ORNL, Wes is leading an international initiative to develop ExaDigiT, an open-source framework for modeling liquid-cooled supercomputers, exemplified by its application on Frontier, an exascale supercomputer. 

Expected Outcomes from the workshop

Attendees should be able to identify and relate to the key characteristics of what comprises a digital twin and their uses in high performance computing scenarios across domains covered.

Attendees should be able to assess potential real-world example use cases and identify the appropriate means and methodologies for representing them in the digital twin world.

By the end of the workshop, attendees will come away with a deeper understanding of the current state of the art of digital twin technologies, with realistic expectations and the means and methods to leverage in their varied scientific domains and existing or new digital twin projects.

An intended outcome is the establishment of an ongoing community of practitioners who will continue to further the development of digital twins in high performance computing environments.