Author: Benny Akesson

What if we could develop the next-generation of cyber-physical systems with all the bells and whistles faster and more cost-effectively than the previous, while still ensuring that they work correctly? Well, the truth is that industrial cyber-physical systems, such as lithography machines, x-ray machines, production printers, radars, and cars, are getting increasingly complex and difficult to efficiently design, implement, and evolve. Addressing this complexity through new design methodologies has been the common denominator throughout my research career, spanning almost two decades and including both fundamental research in academia and applied research in the Dutch high-tech systems industry. I believe that model-based design methodologies are key to addressing complexity and have many years of experience modelling hardware and software at different levels of abstraction and in a variety of formalisms and use them to generate artifacts, such as code or documentation, or analyze and optimize non-functional properties of systems, such as timing behavior or energy consumption. I enjoy the puzzle of looking at industry challenges, and defining and leading the research needed to address them through a combination of fundamental and applied research projects and programs. I also love working with people and organizations to sustainably embed the developed knowledge and technology, both in industry and academia, turning research results into impact and managing complexity step by step.

A Decade at TNO-ESI: My Journey to Principal Scientist at TNO

Posted on by Benny Akesson

Ten years after joining TNO-ESI, I am proud to announce that I have been promoted to Principal Scientist for Model-based System Performance Engineering.

A Principal Scientist at TNO is a senior scientific leader who combines deep expertise with the ability to shape research directions and drive impact. They act as a bridge between science, industry, and strategic partners, ensuring that cutting-edge research is translated into real-world solutions. Principal Scientists are internationally recognized in their field, lead complex and high-impact research programs, and contribute to TNO’s long-term scientific and strategic agenda. In addition, they serve as role models within the organization, mentoring colleagues, inspiring others, and helping develop the next generation of scientists.

My positioning for the role built on a combination of three key pillars: (1) fundamental research, enabled by my appointment as Endowed Professor at the University of Amsterdam; (2) applied research at TNO-ESI in collaboration with industry partners such as Thales, Philips, and ASML; and (3) science management, through my roles as TNO-ESI Science Lead and Science and Technology Manager at unit ICT, Strategy & Policy (ISP). Through this combination, I have developed a strong overview of the innovation chain across technological readiness levels, as well as a clear understanding of how science connects with market and operations. This has enabled me to shape scientific programs and connect academia, research organizations, and industry within my field and beyond.

I am grateful for the recognition expressed in the announcement on the ESI website and look forward to continue my work in this role. Lastly, I would like to take this opportunity to thank past and present peers, managers, and directors that supported me during my journey and believed in me, even when I doubted.

Excited to share that our NWO OTP project DESIRE has been accepted!

Posted on by Benny Akesson

Together with ASML, University of Amsterdam, Leiden University, and TNO-ESI, we are starting a new research project on design-space exploration for complex distributed cyber-physical systems. Thank you ASML and TNO-ESI for contributing to the project, and to Philips, Canon Production Printing, Thales, Vanderlande, Eindhoven University of Technology, and University of Twente for joining the user committee.

Design-space Exploration for Complex Distributed Cyber-Physical Systems (DESIRE) builds on our earlier work in DSE 2.0, extending it toward an advanced holistic and automated approach to exploring alternative hardware platforms, software changes, and mappings — helping engineers answer critical what-if questions on performance and cost in increasingly complex systems. In particular, we will focus on:
1) Capturing realistic system behavior from traces while scaling to industrial systems with partial observability and complex environments
2) Bridging software and hardware characterization to enable model-based exploration of performance across heterogeneous platforms.
3) Handling extremely large, heterogeneous, multi-objective design spaces.

At the same time, it’s great to see how earlier results are already being picked up, matured, and experimented with in practice by TNO-ESI and ASML, closing the loop between academic research and industrial impact.

Looking forward to this next step in the collaboration!

The announcement of the grant from NWO is available here.

Reflections on a PhD Defense: Real-Time Guarantees in the Edge–Cloud Continuum

Posted on by Benny Akesson

Today, I served on the PhD committee of Nasim Samimi, who defended her dissertation titled “Edge-Cloud-Assisted Real-Time Cyber-Physical Systems.” This work addresses the challenge of providing predictable real-time guarantees for cyber-physical systems deployed across the edge–cloud continuum. As CPS increasingly rely on distributed, shared, and dynamic infrastructures, traditional design-time schedulability analysis becomes insufficient. The dissertation proposes a set of online admission control, scheduling, and orchestration techniques that provide per-job deadline guarantees and controlled service degradation under bursty workloads. The main contributions span formal job-level admission control for multicore servers, weakly-hard real-time guarantees using (M,K)-firmness, and practical deployment mechanisms for real-time workloads in Kubernetes-based edge–cloud platforms. Together, these contributions aim to bridge the gap between real-time theory and modern cloud-native practice.

I very much enjoyed reading this dissertation, as it tackles an important problem at the intersection of real-time systems, cyber-physical systems, and edge–cloud computing. I particularly appreciate the combination of solid theoretical foundations in the early chapters with increasingly practical contributions in the later chapters. A great example of this is the KubeDeadline technology, which extends Kubernetes to schedule Linux containers according to the well-known SCHED_DEADLINE policy and guarantee them a portion of CPU bandwidth with bounded latency. This work successfully translates classic real-time scheduling concepts into modern edge-cloud software architectures.

The defense went well, and shortly after the beadle pronounced the end of the defense (“Hora est!”), Nasim became Dr. Samimi. Congratulations on this achievement! It was a pleasure to serve on this committee, and I wish Nasim all the best in her future career.

Literature Study on Performance Prediction

Posted on by Benny Akesson

Today, Jan Przystal presented his literature study Performance Prediction for Microservice-Based Cyber-Physical Systems: A Cross-Domain Literature Review. The study reviews existing methods for predicting performance in Cyber-Physical Systems (CPS) during the early design phase. It focuses on how resource contention, caused by shared hardware such as caches, memory, and I/O, leads to performance degradation, and why this makes Design Space Exploration (DSE) challenging for complex distributed CPS. The study surveys simulation-based approaches, profiling techniques, and interference prediction methods originating from both CPS and cloud computing, and evaluates their suitability for fast and scalable performance estimation. It concludes that while current methods can provide useful approximations, none fully meet the need for a quick, accurate, and scalable prediction approach for large CPS design spaces, highlighting the need for further research in this area. These are challenges that Jan will address during his master project, building on our earlier work with Bruno Dzikowski in this area.

It was nice to see that there was broad interest in Jan’s presentation not only from the performance team at TNO-ESI, but also among some of its industry partners, as well as experts in performance prediction from University of Amsterdam, Eindhoven University of Technology, and University of Twente. That certainly made the Q&A session afterwards spicier, although Jan confidently answered most of the questions.  Congratulations Jan on work well done!

NWO Grants Funding for iCARe Project

Posted on by Benny Akesson

I am pleased to announce that the iCARe project (“Integrated indulgent Control Architecture design”) has been officially granted by NWO under the NXTGEN Hightech programme. The project brings together leading academic and industrial partners to rethink how high‑tech motion systems, such as those used in semiconductor manufacturing, are designed and optimized. With a total project budget of €3.3 million, iCARe aims to develop a radically new integrated control architecture that jointly considers servo control, computational hardware, and power electronics. This approach will enable next‑generation machines to achieve unprecedented accuracy and throughput while remaining cost‑effective, an essential step for future semiconductor technologies.

I will contribute to this project in my role as part-time professor at the University of Amsterdam (UvA). Together with partners from TU/e and ASML, UvA researchers (me, Andy Pimentel, and a PhD student) will develop innovative computational platform architectures, including new scheduling strategies and automated design‑space‑exploration tools that directly link computing performance to control‑system quality. This contribution is vital for enabling high‑precision control at extreme speeds and for integrating computing considerations into the heart of system‑engineering decisions. The project spans six years and will support collaborative research across multiple disciplines.

Congratulations to the iCARE consortium for securing this competitive funding and we look forward to working with you on this next step forward in high‑tech system design.

Read more in the official announcement from NWO or the news at University of Amsterdam.

Call for Special Session Proposals – ESWEEK 2026

Posted on by Benny Akesson

As Special Session Co-chair it is my pleasure to invite Special Session proposals for ESWEEK 2026, the premier event bringing together the embedded systems, software, and cyber-physical systems communities.

Special Sessions are a great opportunity to:
✅ Highlight emerging research directions
✅ Bring together interdisciplinary communities
✅ Foster discussion on timely and impactful topics

We welcome proposals from academia and industry covering innovative, forward-looking, and cross-cutting themes.

📌 Learn more and submit your proposal:
👉 https://esweek.org/call-for-special-session-proposals/
🌐 General information: https://esweek.org

If you are passionate about shaping the technical program of ESWEEK and sparking vibrant discussions, we strongly encourage you to submit a proposal!

Bruno Dzikowski Successfully Defends Master’s Thesis on Performance Prediction

Posted on by Benny Akesson

Yesterday, Bruno Dzikowski successfully defended his master’s thesis titled “Practical Recommendations for Accurately Predicting Performance Degradation Caused by Memory Contention” The thesis addresses the challenge of predicting performance in microservice-based architectures for cyber-physical systems (CPS) running on multi-core platforms, where resource contention significantly impacts accuracy. Existing methods model interference sensitivity and contentiousness but lack practical implementation guidelines.

 

Bruno’s work introduces a compositional performance prediction framework with three key contributions: 1) a validated contentiousness profiling component, 2) an analysis of how system configuration affects prediction accuracy, and 3) the design and implementation of an experimental testbed. Tested across 195 co-location scenarios, the approach achieves high accuracy (median error ≈ 1.4%), demonstrating its effectiveness for forecasting microservice performance.

We are very proud of the excellent research Bruno conducted during his internship with TNO-ESI, which resulted in an outstanding thesis that was confidently presented and defended. We thanks Bruno for the excellent collaboration and wish him all the best for his future career.

Celebrating Dr. Panos Giannakopoulos’ Dissertation Defense

Posted on by Benny Akesson

Congratulations to the newly minted Dr. Panagiotis (Panos) Giannakopoulos, who has successfully defended his dissertation, “Predictable Application Performance in Resource Clusters” The dissertation tackles the challenge of meeting strict Round-Trip Time (RTT) deadlines for time-sensitive applications in heterogeneous, resource-constrained edge environments by developing lightweight, accurate performance predictors. These predictors leverage selected system metrics and machine learning models to anticipate execution time and variability, enabling proactive scheduling and load balancing that improve efficiency and reduce resource waste, with demonstrated success on Electron Microscopy workloads in Kubernetes-based clusters.

This research was conducted as part of the NWO ADAPTOR project, co-funded by Thermo Fisher Scientific and ASTRON. I have had the pleasure of serving on the user committee for this project over the past couple of years and was honored to be invited to join the Ph.D. committee. Over the years, Panos has presented his work at TNO-ESI several times in various settings and was also invited to share his insights at Thales. Panos will now continue his work as a postdoctoral researcher at TU/e.

Hyperheuristic Optimization in Cyber-Physical System Design

Posted on by Benny Akesson

Today, we proudly celebrate Lars van der Water’s successful defense of his Master’s thesis, Exploring Vast Design Spaces with Hyperheuristics: Theoretical Foundations and Autotuning Implementation, at the University of Amsterdam. This work has been conducted in connection with the DSE2.0 project, a research collaboration between University of Amsterdam, Leiden University, and ASML, co-funded by NWO and TNO-ESI as a part of the Mastering Complexity (MasCot) Program.

Lars’ thesis addresses the growing complexity in designing Distributed Cyber-Physical Systems, which are increasingly vital to infrastructure and industry. Traditional Design Space Exploration methods struggle with scalability, algorithm selection, and parameter tuning, creating a bottleneck in efficient exploration of system designs. To overcome this, this work explores hyperheuristics (HHs) as a higher-level domain-agnostic approach to automate the selection and tuning of metaheuristics.  Key contributions include a modular framework for integrating HH strategies, and empirical insights into the trade-offs between performance, effort, and computational cost in autotuning. Experiments show promising results for auto-tuning of simpler meta-heuristic search algorithms like Gravitational Search and Particle Swarm Optimization, but revealing limitations with more complex ones like Genetic Algorithms.

We sincerely thank Lars for the excellent collaboration and wish him all the best in the next chapter of his career!

Paper on Parallelism in dCPS Workload Modeling Accepted at Euromicro DSD 2025!

Posted on by Benny Akesson

I am thrilled to announce that the paper “Unraveling Parallelism in Automated Workload Modeling for Distributed Cyber-Physical Systems” has been accepted for publication at the 28th Euromicro Conference Series on Digital System Design (DSD). This paper was first-authored by Faezeh Sadat Saadatmand and is a result from the DSE2.0 project, a collaboration between University of Amsterdam, Leiden University, and ASML.

The paper addresses the problem of limited exploration of software-level parallelism in distributed Cyber-Physical Systems, due to fixed execution orders in current workload models used in design-space exploration. It proposes refined workload models based on execution traces that capture both inter- and intra-process dependencies, enabling safe task reordering and parallel execution without modifying the software. A case study on the ASML Twinscan lithography machine demonstrates performance improvements while maintaining functional correctness.