Keynote Speakers
Ilkay Altintas
University of California San Diego, USA
Societal Computing and Innovation in the AI Era
Hubertus Franke
IBM T.J. Watson Research Center, USA
Data Confidentiality in the World of Agentic AI Systems
Julie A. McCann
Imperial College London, UK
Rubies in the Dust
Anne-Cécile Orgerie
IRISA Rennes, CNRS, France
Carbon Footprint Allocation Models in Distributed Systems
Anne-Cécile Orgerie
IRISA Rennes, CNRS, France
Carbon Footprint Allocation Models in Distributed Systems
Distributed systems are increasingly spanning worldwide, with digital services hosted all around the globe and often belonging to complex systems, utilizing many other services and hardware resources themselves. Along with this increase comes an alarming growth of energy consumption and carbon footprint. Despite the distributed systems’ complexity, understanding how they consume energy is important in order to hunt wasted Joules and reduce their environmental impact. This talk will deal with attributing carbon footprint to users of distributed systems.
Dr. Anne-Cécile Orgerie is a permanent research scientist (Directrice de recherche) at CNRS (Centre National de la Recherche Scientifique), working in the Magellan team at IRISA in Rennes, France. She was formerly a postdoctoral researcher at the Department of Electrical and Electronic Engineering at the University of Melbourne, Australia. Dr. Orgerie obtained her PhD from ENS de Lyon, France, in September 2011. Her main research focuses on the energy efficiency and environmental impacts of distributed systems, including cloud infrastructures, telecommunication networks, and smart grids.
Julie A. McCann
Imperial College London, UK
Rubies in the Dust
Over the past 25 years I have been studying wireless sensing systems and their evolution from dumb embedded computation to intelligent edge and IoT systems. This type of computation is highly distributed, decentralised, and is prone to effects not found in more general computing (e.g. in data centres). However, I find this interesting, and look specifically at this cyber-physical interaction to find ways around problems, optimise systems or reuse our infrastructures – essentially turning noise into new or better signals – therefore finding Rubies in the Dust.
Julie A. McCann is a Professor of Computer Systems with Imperial College London and is currently Co-Director of the School of Convergence Science in Security, Space and Telecoms and Director of the national CHEDDAR communications research hub. Formerly Vice Dean Research in the Faculty of Engineering, she has published extensively on decentralized and self-organizing scalable algorithms and protocols for Wireless/RF Sensor-based systems, Internet of Things, and Cyber-physical systems. She leads the Adaptive Emergent Systems Engineering Research (AESE) research group, and between 2015-2022 was the Deputy Director of PETRAS IoT Cybersecurity Hub, Critical Ecosystems Lead for the Alan Turing Institute, and Imperial PI on the EPSRC programme grant Science for Sensor Systems Software. She has a number of international research collaborations including Singapore NRF funded Eco-Cities (until March 2024 she had a sub-lab in Singapore with I2R and HDB), between 2012-2017 directed the Intel Collaborative Research Institute (ICRI) for Sustainable Cities, as well as other projects through EU FP7/H2020 programmes. McCann is an elected Member of the Council of Computer Science Professors and Heads of Computing, and was elected to the membership committee of the UKCRC. She holds the 2018 UKRI Suffrage Science Award for Computing and Mathematics, President’s Medal for Research Excellence 2020, and is a Fellow of the BCS and Chartered Engineer.
Hubertus Franke
IBM T.J. Watson Research Center, USA
Data Confidentiality in the World of Agentic AI Systems
Over the last two decades, the fusion of cloud computing, edge computing, and artificial intelligence has revolutionized how we build and operate digital systems. This convergence has enabled data and computation to flow fluidly across enterprise backends, cloud platforms, edge infrastructure, and endpoint devices—creating a highly distributed and intelligent ecosystem. In this new landscape, data has become a strategic asset—valuable, sensitive, and increasingly exposed to risk. While encryption of data at rest and in transit has become standard, it does not protect data during active use. This gap has led to the rise of confidential computing, a paradigm that uses hardware-based memory encryption and secure enclaves to protect data and code during execution, even in potentially compromised environments. Confidential computing has matured rapidly, demonstrating that many existing applications can run securely within enclaves with minimal changes. But as we move into the era of agentic AI systems—autonomous agents capable of making decisions, initiating actions, and learning across diverse platforms—the need for robust, end-to-end security becomes even more urgent. These systems often operate without direct human oversight, processing sensitive data and interacting with users in real time. This raises critical concerns about privacy, model integrity, and trust—especially in the context of large language models (LLMs) and other AI-driven services. In this keynote, we argue that confidential computing should serve as the security foundation for agentic AI. It offers the trust guarantees necessary to protect both user data and model behavior in dynamic, distributed environments. However, to fully realize this vision, AI accelerators—such as GPUs and specialized inference chips—must evolve to support enclave-aware execution, ensuring that performance optimizations do not compromise security. As organizations increasingly rely on AI to drive innovation, confidential computing offers a critical pathway to building secure, intelligent, and scalable systems. In a world where data is central and trust is non-negotiable, it provides the infrastructure needed to ensure privacy, compliance, and resilience in the age of autonomous intelligence.
Dr. Hubertus Franke is a Distinguished Research Scientist at the IBM T.J. Watson Research Center. He received a Diplom Informatik degree from the Technical University in Karlsruhe (KIT), Germany (1987) and a Ph.D. in Electrical Engineering from Vanderbilt University, USA (1992). Since joining IBM Research in 1993, Dr. Franke has worked in many areas of the system stack. He was responsible for the ASCI-supercomputing MPI messaging and gang scheduling system. He has extensive experience in Operating Systems being responsible for the Enterprise Linux Research strategy in the early 2000s with several contributions to the Linux kernel. He was a lead architect for the PowerEN architecture and in particular the novel on-chip accelerator integration. He was senior manager of the software-defined systems and cloud department working on networking, scheduling, and infrastructure management, and most recently he works on confidential computing and AI integration. Dr. Franke is an ACM Fellow (Association for Computing Machinery), an IBM Master Inventor (184 patents) and is widely published (150+). He is an Adjunct Full Professor at NYU (2011-), an Adjunct Research Professor at UIUC (2022-) and an Adjunct Professor at Columbia (2024-).
Ilkay Altintas
University of California San Diego, USA
Societal Computing and Innovation in the AI Era