Mariano Scazzariello is a Lecturer at KTH Royal Institute of Technology, Sweden, affiliated with the School of Electrical Engineering and Computer Science and the Department of Network and Systems Engineering. He teaches the course 'Network Systems with Edge or Cloud Datacenters (IK2227)'. His research focuses on advanced networking topics including machine learning in networks, high-speed packet processing, network emulation, and software-defined networking innovations. His work spans contributions to network emulation tools like Kathará and Megalos, stateful packet processing at terabit scales, and leveraging large language models (LLMs) for network configuration and vulnerability detection. Recent research emphasizes low-latency protocols (e.g., SRv6/DetNet integration) and GPU-centric networking on commodity hardware. Mariano’s publications (2020–2025) highlight expertise in network function virtualization, ASIC-based switching, and optimizing network configurations through AI-driven approaches. He has pioneered frameworks for evaluating routing protocols and virtualizing large network scenarios at scale.
Zhonghai Lu is a Professor of Electronic Systems Design (specializing in Dependable and Autonomous Systems) at KTH Royal Institute of Technology, part of the Department of Electrical Engineering in the School of Electrical Engineering and Computer Science (EECS). He serves as Program Director for KTH's Embedded Systems master's program and Director of Studies at the Division of Electronics and Embedded Systems. His research focuses on Network-on-Chip (NoC), computer architecture, embedded systems, and Prognostics and Health Management (PHM) of power electronics. He leads a research group exploring in-network processing and embedded intelligence, transforming passive networks into active computational frameworks. Lu holds a BSc from Beijing Normal University (1989), MSc and PhD from KTH (2002, 2007), and an MBA in Innovation and Growth from the University of Turku (2012). He has authored over 240 scientific papers, including journal articles and peer-reviewed conferences, with notable recognitions such as Best Paper Awards at NOCS’2015 and EU HiPEAC, and a Featured Paper in IEEE Transactions on Computers (2020). He serves as Associate Editor for ACM Transactions on Architecture and Code Optimization (TACO) and has chaired major conferences like HiPEAC’2017 and NOCS’2018. His research group’s recent work includes integrating AI into hardware acceleration, fault-tolerant neural networks, and RUL estimation for power electronics using recurrent neural networks. Lu has secured grants from the Swedish Research Council and Intel Corporation and developed courses like IL2230 (Hardware Architectures for Deep Learning) and IL2233 (Embedded Intelligence), pioneering embedded AI education at KTH. Education: BSc (Beijing Normal University), MSc/PhD (KTH), MBA (University of Turku) Awards: Best Paper Awards (NOCS, EU HiPEAC), Swedish Research Council Grants, Intel Research Gifts Labs/Teams: Research Group on In-Network Processing and Embedded Intelligence
Stefano Markidis is a Professor of Computer Science at KTH Royal Institute of Technology, affiliated with the School of Electrical Engineering and Computer Science and the Digital Futures Faculty. He holds a Ph.D. from the University of Illinois at Urbana-Champaign and an MS from Politecnico di Torino. His research focuses on high-performance computing systems, including supercomputers and quantum computers, with expertise in plasma simulations, quantum algorithms, and scalable computational frameworks. Markidis leads the development of the Neko framework for high-fidelity computational fluid dynamics and the iPIC3D particle-in-cell code for plasma physics. He teaches courses such as Quantum Computing for Computer Scientists, High-Performance Computing, and Applied GPU Programming. His work addresses exascale computing challenges, including optimizing algorithms for GPUs, quantum systems, and distributed architectures. Key research interests include: Parallel Programming Models and HPC Frameworks Quantum Computing Applications in Scientific Simulations Physics-Informed Machine Learning Exascale System Optimization Turbulence Modeling and Plasma Dynamics His publications span over 100 articles in journals like Journal of Computational Physics and Scientific Reports , focusing on topics such as scalable CFD, quantum neural networks, and plasma simulation techniques. He has advised numerous students in these areas. Markidis collaborates with institutions like Los Alamos National Laboratory and RISE Research Institutes of Sweden through the Digital Futures initiative, aiming to solve societal challenges via digital technologies.
Mikael Johansson is a Professor at Kungliga Tekniska Högskolan (KTH), specializing in Control Technology . He teaches and coordinates courses such as Distributed Optimization (FEL3311) and various advanced-level degree projects in computer science, electrical engineering, and systems engineering. His research spans Control Systems , Machine Learning , and Optimization , with a focus on asynchronous algorithms, federated learning, and applications in energy systems and construction. His work includes 15 recent publications on topics like neural networks, distributed optimization, and battery technology. Notable areas of contribution are in asynchronous learning, federated learning with privacy constraints, and quasi-Newton methods for optimization. His research bridges theoretical advancements with practical applications in urban design, healthcare, and autonomous systems.
Anders Söderholm serves as the Vice-Chancellor (Rektor) of KTH Royal Institute of Technology, Sweden's leading technical university. His leadership spans academic, research, and strategic domains with a strong focus on positioning KTH as a key player in addressing global challenges through technology and innovation. Recent initiatives highlight his emphasis on AI ethics, quantum technology development, and sustainability research. Söderholm's research interests center on academic leadership, organizational theory, and project management, with particular focus on temporary organizations and the 'projectified society.' His work bridges theoretical frameworks with practical applications in higher education governance and research policy. He has significantly contributed to the Scandinavian school of project management research, exploring how organizations navigate complexity through temporary structures. Analysis of his recent publications reveals a clear trajectory from theoretical explorations of project management toward practical applications in higher education leadership. His scholarship demonstrates increasing engagement with the challenges of university governance in an era of rapid technological change, particularly around AI implementation and international research collaboration. The interdisciplinary nature of his work connects organizational theory with practical leadership challenges in academic settings. Söderholm actively engages in national policy discussions, having contributed to debates about university associations, research funding models, and international collaborations, particularly regarding China. His leadership extends to fostering partnerships with industry and other academic institutions, as evidenced by initiatives with Chalmers University of Technology. As Vice-Chancellor, Söderholm oversees KTH's strategic direction, which includes significant investments in research infrastructure, educational innovation, and international collaboration. His leadership approach emphasizes the university's role in societal development, technological advancement, and democratic values, while navigating the complex landscape of research funding and academic autonomy.
Lars Davidson is a Professor in the Department of Fluid Dynamics at Chalmers University of Technology. His research focuses on numerical simulations of fluid flow and heat transfer, with an emphasis on turbulence modeling for Large Eddy Simulation (LES) and hybrid LES/RANS methods. He has developed computational codes CALC-BFC and CALC-LES based on finite-volume techniques, and recently integrated machine learning to enhance wall functions and turbulence models. Key projects include Hybrid LES/RANS for wall-bounded flows Machine learning applications in fluid dynamics Aeroacoustic noise reduction in automotive and aerospace systems Wind turbine load analysis in forested regions . His publications span 302 articles in journals and conferences, with recent work on Neural networks for turbulence closure Plasma actuators for drag reduction Lattice Boltzmann wall-modeled LES . Collaborations include teams at Volvo, Siemens, and international research groups.
Zebo Peng is a Professor and Deputy Head of Department at Linköping University's Department of Computer and Information Science (IDA), leading the Software and Systems (SAS) division. His research focuses on embedded systems design, electronic design automation, SoC testing, and real-time systems with emphasis on fault tolerance and hardware/software co-design. He has contributed to projects like the ASTECC initiative, funded by the Swedish Foundation for Strategic Research, addressing adaptive software in edge-cloud continuum systems. Key research interests include cyber-physical systems security, time-sensitive networking (TSN), and optimization techniques using genetic algorithms. Recent work explores thermal-aware design for reliability, security-aware scheduling, and stability guarantees in control systems. His publications span journals like IEEE TPDS and ACM TECS, alongside conference contributions on topics like resource management and fault detection in distributed systems. Prof. Peng collaborates extensively within the SAS division, which bridges academic and industrial research in software engineering and computer systems. His team's projects address challenges in real-time systems, embedded security, and parallel computing architectures.
Professor Per Stenström is affiliated with the Department of Computer Science and Engineering at Chalmers University of Technology . His research focuses on computer architecture , memory systems optimization , and energy-efficient computing , with significant contributions to DNN accelerator design and cache management . Research Trends : His recent publications emphasize Memory compression techniques for energy efficiency Hardware-software co-design for DNN acceleration Security in microarchitectural optimizations Hybrid memory systems for near-memory computing These works span both theoretical and applied aspects of computer architecture, with a particular focus on data redundancy elimination , parallel processing , and quality-of-service constraints . His work has influenced the development of energy-aware resource management frameworks and resilient EU HPC systems , as evidenced by his long-standing contributions to the field since the early 2010s.
Philippas Tsigas is a Professor at the Department of Computer Science and Engineering at Chalmers University of Technology. He leads the Distributed Computing and Systems Research Group and has held roles as co-leader of research initiatives such as the PEPPHER project. His research spans distributed/parallel computing, information visualization, and fault-tolerant communication mechanisms. He has supervised numerous PhD students, including Yi Zhang, Håkan Sundell, and Farnaz Moradi. Research interests include lock-free data structures, multicore algorithms, secure network services, and visualization tools like Lydian and DataMeadow. Notable awards include Best Paper Awards at IPDPS 2003 and SNS 2012. His work has been published in top venues like IEEE Transactions on Parallel and Distributed Systems and ACM Journal of Experimental Algorithmics. Awards highlight contributions to lock-free algorithms and network modeling. Students have contributed to projects like NBmalloc (memory reclamation) and GPU Quicksort. Collaborations with institutions like SSF and VR have supported his research. Tsigas is also involved in teaching distributed systems and mentoring early-career researchers.
Niclas Jansson is a researcher at the PDC Center for High Performance Computing at KTH Royal Institute of Technology. He holds an M.S. in Computer Science (2008) and a Ph.D. in Numerical Analysis (2013) from KTH. His career spans roles such as postdoctoral researcher at RIKEN Advanced Institute for Computational Science (2013-2016) and visiting scientist at RIKEN (2018-2021), where he contributed to the Japanese exascale program Flagship 2020. A core focus of his research involves extreme-scale computing and numerical method development. He is a key developer of RIKEN's multiphysics framework CUBE , the HPC branch of FEniCS , and the spectral element flow solver Neko . His work is currently supported by a Swedish Research Council Starting Grant aimed at enhancing high-order spectral element methods for exascale fluid simulations. Niclas has published extensively on topics such as GPU acceleration , adaptive finite element methods , in situ visualization , and extreme-scale turbulence modeling . He also teaches Computational Fluid Dynamics (SG2212) at KTH.
Marjan Firouznia is a Principal Research Engineer at Linköping University , affiliated with the Division of Diagnostics and Specialist Medicine (DISP) under the Faculty of Medicine and Health Sciences . With a PhD in Electrical Engineering from Amirkabir University of Technology and postdoctoral experience at institutions like Case Western Reserve University, she specializes in advancing machine learning models for precise segmentation of cardiac structures including the left atrium , epicardial fat , and fibrosis using CT and MRI scans. Her work aims to improve diagnostic accuracy and treatment planning in cardiovascular care. Marjan's research focuses on medical imaging , deep learning , and computational anatomy , with recent publications on FractalRG , FK-means , and Poincare-guided UNet for cardiac structure segmentation. Her academic contributions span 15 recent publications , emphasizing fractal geometry , chaos theory , and optimization algorithms in biomedical applications. She actively develops open-source datasets and tools, such as the FK-means codebase , to support reproducibility in medical AI research.
Christoph Kessler is a Professor and Head of the Software and Systems (SAS) division at the Department of Computer and Information Science (IDA), Linköping University, Sweden. He leads the Programming Environment Laboratory’s research group focusing on compiler technology, parallel computing, and heterogeneous systems. His work includes the development of tools like OPTIMIST, PARAMAT, and SkePU, and he has contributed over 100 publications in journals and conferences. He holds a PhD from the University of Saarbrücken and a Habilitation from the University of Trier. Research interests span parallel programming, compiler optimization, and energy-efficient scheduling for heterogeneous systems. He has secured a 30M SEK grant from SSF for the ASTECC project, advancing adaptive software for edge-cloud computing. Notable contributions include frameworks for GPU-based systems and methodologies for optimizing resource allocation on many-core architectures. His team’s work emphasizes practical applications in high-performance computing, including tools for course management (StASy) and energy-aware scheduling algorithms. The SAS division, under his leadership, focuses on software engineering and computer systems research with strong industry collaboration.
Garrelt Mellema is a Professor in the Department of Astronomy at Stockholm University, specializing in computational astrophysics and cosmology. His research focuses on the Epoch of Reionization, the period when the first stars and galaxies formed approximately 13 billion years ago. He leads work in developing computational tools for astrophysical research across various domains from solar physics to cosmology. Professor Mellema's primary research interest centers on the Epoch of Reionization and Cosmic Dawn, particularly studying the 21-cm signal from neutral hydrogen. His work employs advanced computational methods including radiative transfer simulations (C2-Ray, pyC2Ray), machine learning techniques, and analysis of observational data from radio telescopes like LOFAR and the future SKA. His research group develops computational tools for studying cosmic reionization, the formation of the first structures, and the evolution of the intergalactic medium. The analysis of his recent publications reveals a strong focus on extracting the faint 21-cm signal from observational data using innovative techniques including neural networks and advanced statistical methods. His work bridges theoretical modeling with observational constraints, particularly from LOFAR observations, to understand the physical conditions during the cosmic dawn and epoch of reionization. Current research trends show increasing integration of machine learning with traditional astrophysical methods to overcome systematic challenges in 21-cm cosmology. As leader of the Computational Astrophysics Group at Stockholm University, Professor Mellema oversees development of simulation tools used by the international community studying cosmic reionization. His work on the C2-Ray radiative transfer code has become a standard tool in the field, with GPU-accelerated versions enabling more detailed simulations of the complex processes during the formation of the first luminous objects in the universe.
Yuan Yao serves as an Assistant Professor in the Department of Information Technology at Uppsala University, Sweden. His academic role spans teaching and research within the Computer Systems division, focusing on cutting-edge computer architecture and parallel computing systems. He maintains active collaborations across international institutions, particularly in energy-efficient hardware design and emerging computing paradigms. His educational journey includes: B.S. in Micro-electronics from Northwestern Polytechnical University, China (2009) M.S. in System-on-Chip Design from KTH Royal Institute of Technology, Sweden (2014) Ph.D. in Electrical Engineering and Computer Science from KTH Royal Institute of Technology (2019) Yao's research centers on power and thermal management for chip multi-processors, Network-on-Chips (NoCs), and GPUs. He pioneers hardware/software co-design for high-performance computing, coherency mechanisms for emerging memory technologies, and performance analysis of on-chip networks. Recent work expands into neural network acceleration and battery-less Internet of Things architectures, reflecting a trajectory toward energy-constrained specialized systems. His methodology integrates formal modeling with practical implementation for real-world impact. Publication trends reveal consistent innovation in energy efficiency across parallel architectures. From foundational DVFS techniques for NoCs (2016-2018) to recent breakthroughs in battery-less IoT (2023-2024), his work demonstrates evolutionary progression toward novel computing domains. Key thematic threads include thermal-aware optimization, memory consistency protocols, and hardware acceleration for AI workloads, with applications spanning data centers to embedded systems. Scientific recognition includes: Best paper candidate at IEEE International Symposium on High Performance Computer Architecture (HPCA) 2018 for in-network packet generation research Yao actively supervises graduate researchers and leads collaborative projects in computer architecture. His grant portfolio supports work on battery-less IoT systems and neural network accelerators, though specific funding details aren't publicly enumerated. Current projects emphasize sustainable computing through novel architectures for energy-harvesting environments. He operates within Uppsala University's Computer Systems division, contributing to research groups focused on hardware acceleration, embedded systems, and networked architectures. His lab environment fosters interdisciplinary work bridging computer architecture, energy harvesting, and machine learning for next-generation computing platforms.
Seif Haridi is a Professor at KTH Royal Institute of Technology in Stockholm, Sweden, specializing in parallel and distributed computing systems. He holds dual roles as Chair-Professor of Computer Systems and Chief Scientific Advisor at RISE SICS. His research integrates systems engineering with theoretical foundations, focusing on programming systems, distributed computing, and big data technologies. Key contributions include co-designing SICStus Prolog, the Mozart Programming System, and Apache Flink, as well as leading the development of HOPS, a European big data platform awarded the IEEE Scale Prize 2017. He has led major EU projects like EIT-Digital’s cloud computing initiative and co-founded startups such as LogicalClocks and HiveStreaming. His teaching includes courses on distributed algorithms and peer-to-peer computing at KTH. Notable awards include the European Data Science Technology Innovation 2019. His work spans systems like HOPS, Flink, and Kompics, emphasizing scalability and robustness in distributed environments. Current projects include CDA (Continuous Deep Analytics) and ExtremeEarth for geospatial data analysis. Research interests include distributed algorithms, consensus protocols, and cloud-native systems. His lab’s contributions to scalable storage (e.g., HopsFS) and stream processing (Apache Flink) highlight his impact on both academia and industry.
