Keynote speakers
Angelo Corsaro: CEO, CTO, and co-founder of ZettaScale Technology, formerly a founding Co-Chair of the OMG DDS standard.
Title: Zenoh: Unconstrained Communication, Compute and Storage from the Cloud to the Microcontroller.
Abstract: Modern systems—ranging from mobile robots and aerial drones to connected and autonomous vehicles—seamlessly span across the compute continuum, from microcontrollers to large-scale data centers. Yet, much of today’s technology stack remains fragmented, relying on legacy solutions designed for a vastly different era. The result? Inefficient, architecturally disjointed, and energy-draining systems—the closest approximation to what one might call a Digital Frankenstein.
This challenge is exacerbated by the proliferation of battery-powered devices, where energy efficiency is paramount. Inefficient communication protocols and rigid storage policies force unnecessary data transfers, consuming both network resources and precious battery life. Addressing these inefficiencies requires a fundamental shift in how we approach communication, computation, and storage—not just incremental patches on outdated paradigms.
In this keynote, we will trace the evolution of distributed computing middleware, exploring the key technological drivers that have shaped it over the years. We will then introduce Zenoh, a next-generation distributed computing framework that unifies communication, computation, and storage across the entire compute continuum, from microcontrollers to the cloud. We will examine why Zenoh is emerging as the de facto protocol for next-generation robotics, intelligent transportation systems, and advanced IoT/Edge applications. Finally, we will look ahead to the future of distributed computing, discussing upcoming trends and how they will shape the next wave of intelligent, energy-efficient systems.
Bio: Angelo Corsaro, Ph.D., is the Chief Executive Officer (CEO), Chief Technology Officer (CTO), and co-founder of ZettaScale Technology. At ZettaScale, he leads a world-class team dedicated to revolutionizing distributed computing, ensuring that every connected human and machine can communicate, compute, and store data anywhere, at any scale, efficiently, and securely.
A globally recognized expert in Cloud-to-Thing Continuum, high-performance distributed systems, and real-time computing, Angelo is the inventor of the Zenoh protocol and the lead of the Eclipse Zenoh project. His work is reshaping the future of data-centric and event-driven architectures, bridging the gap between cloud, edge, and embedded systems.
With a prolific research background, Angelo has authored over 100 publications in peer-reviewed journals, conferences, workshops, and industry magazines. He has also contributed to over ten international standards, shaping the evolution of real-time and large-scale distributed computing.
Angelo was a founding Co-Chair of the OMG DDS standard, playing a key role in its development and adoption. His leadership in DDS continued until 2015, when he stepped away due to technical divergences and committed himself to designing and implementing the first Zenoh protocol, paving the way for the next generation of data-centric middleware.
Mohammad Al Faruque: Full professor at the University of California Irvine and Director of the Cyber-Physical Systems Lab.
Title: Cyber-Physical Vulnerabilities in Autonomous Systems: Facing Hard Realities and Shaping the Road Ahead!
Abstract: Cyber-physical systems (CPS) are in every security-sensitive autonomous system, such as aerospace, automotive, UAVs, robotics, energy, healthcare, manufacturing, transportation, and consumer appliances. CPSs are engineered systems that are built from and depend upon the seamless integration of computation and physical components [NSF]. Embedded systems comprising hardware and software systems are the primary enabling technology for these cyber-physical systems. Compared to traditional information processing systems, new vulnerabilities emerge from the boundaries between various layers due to the tight interactions between cyber and physical components in CPSs and closed-loop control from sensing to actuation. In this Keynote, Prof. Al Faruque will discuss how new vulnerabilities emerge at the intersection of various components and subsystems and their various hardware, software, communication, and physical layers. Several recent examples from various cyber-physical systems will be presented in this talk. A very different set of methodologies and tools are needed to understand these new vulnerabilities. Defenses against these vulnerabilities also demand new hardware/software co-design approaches. The keynote will highlight recent developments in this regard. The primary goal of this keynote will be to highlight various research challenges and the need for novel scientific solutions from the larger research community.
Bio: Mohammad Al Faruque is currently with the University of California Irvine (UCI), where he is a full professor and directs the Cyber-Physical Systems Lab. He is currently the ACM Distinguished Speaker and was the IEEE CEDA Distinguished Lecturer for 2022-2023. Prof. Al Faruque also directs the Samueli School of Engineering Autonomous Systems Initiatives. Among many awards, Prof. Al Faruque received 4 best paper awards, including ACSAC, DAC, DATE, and ICCAD. He received many best paper award nominations from the top embedded and cyber-physical systems conferences. He also received the IEEE Technical Committee on Cyber-Physical Systems Early-Career Award and the IEEE CEDA Ernest S. Kuh Early Career Award. Besides 175+ IEEE/ACM publications in the premier journals and conferences, Prof. Al Faruque holds 12 US patents. He has been awarded the Thomas Alva Edison Patent Award for one of his inventions.
Matteo Andreozzi, Senior Technology Leader at Arm
Title: Architecting the Future: Scalable QoS and Predictability with MPAM in Heterogeneous Systems
Abstract: In this presentation, the author explores the innovative Memory Partitioning and Monitoring Architecture (MPAM) developed by Arm.
The MPAM architecture is designed to enhance the performance and predictability of heterogeneous systems by providing scalable Quality of Service (QoS) and resource partitioning.
The presentation delves into the principles and mechanisms of MPAM, including the generation and control of labels, resource partitioning, and monitoring. It highlights the practical applications of MPAM in various domains such as autonomous machines, enterprise infrastructure, and smart devices.
The presentation also addresses the challenges of resource contention and performance interference in shared resources, offering insights into how MPAM mitigates these issues to ensure consistent and reliable system performance.
Through this comprehensive overview, the presentation aims to provide a deep understanding of how MPAM can be leveraged to architect the future of scalable and predictable heterogeneous systems.
Bio: Matteo Andreozzi is a senior technology leader at Arm, where he spearheads the development of next-generation computing architectures with a focus on high-performance heterogenous systems. He serves as Lead Architect for the Memory Partitioning and Monitoring Architecture, guiding efforts to ensure Arm-based systems efficiently manage the complexity of memory resource allocation, contention, and interference in domains like mobile devices, automotive platforms, robotics, and cloud/infrastructure systems.
Andreozzi’s expertise lies in areas such as quality-of-service (QoS) management, deterministic real-time computing, and system performance optimization.
With over 15 years of experience bridging academic research and industry – including prior work at NVIDIA and extensive research during his Ph.D. – he has made significant contributions to both experimental research and practical product development in the field of computer architecture.
Alberto Brandolini, EventStorming and Avanscoperta
LLM-Agent Architectures: blueprinting the Next Wave of Autonomous Software
Abstract:
Large Language Models (LLMs) are no longer just powerful text generators—they are becoming autonomous agents capable of perception, reasoning, planning, and action. This keynote explores the architectural shift driven by LLM-agents, examining how they challenge our software engineering assumptions, development processes, and system designs. Drawing from recent research on agentic workflows, modular architectures, and design patterns in intelligent systems, we will explore how LLM-agents are reconfiguring the boundaries between humans, software, and tasks.
Through real-world examples, emerging blueprints, and critical reflections, we will unpack both the promise and the pitfalls of LLM-agents. We will reflect on key architectural decisions, discuss state-of-the-art tools and methods, and imagine what a future software engineering discipline—built for and with agents—might look like.
Short Bio:
Hi, I’m Henry Muccini, a Professor of Software Engineering at the University of L’Aquila in Italy. I earned my Ph.D. in Computer Science from the University of Rome “La Sapienza,” followed by a postdoc at the University of California, Irvine.
Over the years, I’ve seen software architecture evolve in fascinating ways—and lately, I’ve been especially drawn to how AI, large language models (LLMs), and LLM-based agents are changing how we build and reason about complex software systems. I’ve been part of the software architecture community for more than two decades and currently serve as Chair of the Steering Committee for ICSA (the IEEE International Conference on Software Architecture), and a member of the Steering Committees for ECSA (the European Conference on Software Architecture) and CAIN (the IEEE Conference on AI Engineering – Software Engineering for AI).
I’m also the Associate Editor in Chief of IEEE Software, where I get to look at how the latest technologies—like LLM agents—are making their way into real-world software engineering practice.
Since 2019, I’ve been working at the intersection of software architecture and machine learning, through papers, workshops, special issues, Dagstuhl seminars, and more. I’m always eager to share what I’ve learned and learn from others in return.
To round out my introduction: I currently lead four international double-degree programs in software engineering, with a focus on machine learning and sustainable ICT. I also head the FrAmeLab research laboratory, where we explore emerging challenges at the intersection of AI, green software, and software systems. Previously, I served as Head of the Computer Science program at the University of L’Aquila.
Feel free to connect with me on LinkedIn or visit the FrAmeLab website to learn more.