Tomasz Szydlo
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Tomasz Szydlo has a total of 112 co-authors. Here are the top co-authors by number of shared publications:
Robert Brzoza-Woch
Institute of Computer Science, AGH University of Science and Technology
17 shared publications- Rajiv Ranjan
Newcastle University
13 shared publications - Devki Nandan Jha
Newcastle University, Newcastle upon Tyne, United Kingdom
8 shared publications Krzysztof Zielinski
AGH Univ Sci & Technol, Dept Comp Sci, Krakow, Poland
8 shared publicationsMarek Konieczny
AGH University of Science and Technology Faculty of Computer Science, Electronics and Telecommunications Department of Computer Science Al. Mickiewicza 30, 30-059 Kraków, Poland
6 shared publicationsJoanna Sendorek
Institute of Computer Science, AGH University of Science and Technology
5 shared publications
Latest Publications
- 12 October 2026
- Published by Springer Nature
- p. 247-268
Big Data Processing in the IoT-Edge-Cloud Continuum
- 12 October 2026
- Published by Springer Nature
- p. 247-268
Big Data Processing in the IoT-Edge-Cloud Continuum
- Open Access
- 3 July 2026
- Published by Elsevier in SSRN Electronic Journal
Towards a Federated Edge–HPC Continuum: Taxonomy, Workflow Systems, Applications, and Research Challenges
Cloud-centric architectures have evolved to support data-intensive and latency-sensitive applications, leading to the emergence of the Edge–cloud continuum, where computation spans devices, edge nodes, and cloud data centres. However, many modern workloads require computational capabilities beyond what edge and cloud infrastructures alone can provide, motivating the integration of High-Performance Computing (HPC) as a complementary layer. This shift introduces the concept of Edge–HPC federation, in which independently managed edge and HPC systems collaborate across administrative and organisational boundaries. In this paper, we analyse the evolution of distributed computing infrastructures and position Edge–HPC federation within the broader context of grid, cloud, and edge paradigms. We introduce a multi-dimensional taxonomy of Edge–HPC workflows, capturing key aspects such as data-flow direction, processing patterns, time sensitivity, data characteristics, workflow types, and security posture. Using this taxonomy, we systematically review existing workflow management systems and map them to emerging application domains, including smart infrastructure, healthcare, environmental monitoring, and scientific experimentation. Our analysis highlights fundamental gaps in current systems, particularly in cross-domain orchestration, data management, and security in federated environments. We further identify key research challenges and outline future directions for enabling seamless, secure, and efficient integration across the Edge-HPC continuum. This work provides a structured foundation for designing next-generation distributed systems that combine real-time edge intelligence with large-scale computational capabilities.
- 3 July 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessTowards a Federated Edge–HPC Continuum: Taxonomy, Workflow Systems, Applications, and Research Challenges
Cloud-centric architectures have evolved to support data-intensive and latency-sensitive applications, leading to the emergence of the Edge–cloud continuum, where computation spans devices, edge nodes, and cloud data centres. However, many modern workloads require computational capabilities beyond what edge and cloud infrastructures alone can provide, motivating the integration of High-Performance Computing (HPC) as a complementary layer. This shift introduces the concept of Edge–HPC federation, in which independently managed edge and HPC systems collaborate across administrative and organisational boundaries. In this paper, we analyse the evolution of distributed computing infrastructures and position Edge–HPC federation within the broader context of grid, cloud, and edge paradigms. We introduce a multi-dimensional taxonomy of Edge–HPC workflows, capturing key aspects such as data-flow direction, processing patterns, time sensitivity, data characteristics, workflow types, and security posture. Using this taxonomy, we systematically review existing workflow management systems and map them to emerging application domains, including smart infrastructure, healthcare, environmental monitoring, and scientific experimentation. Our analysis highlights fundamental gaps in current systems, particularly in cross-domain orchestration, data management, and security in federated environments. We further identify key research challenges and outline future directions for enabling seamless, secure, and efficient integration across the Edge-HPC continuum. This work provides a structured foundation for designing next-generation distributed systems that combine real-time edge intelligence with large-scale computational capabilities.
- Open Access
- 20 June 2026
- Published by Association for Computing Machinery (ACM)
- p. 39-40
Poster: Reference-Based Prompting for Acoustic Anomaly Detection
- 20 June 2026
- Published by Association for Computing Machinery (ACM)
- p. 39-40
Open AccessPoster: Reference-Based Prompting for Acoustic Anomaly Detection
- Open Access
- 20 June 2026
- Published by Association for Computing Machinery (ACM)
- p. 13-18
Juggler: Accuracy v/s Efficiency - Close View on Heterogeneous Embedded Edge ML Serving
The proliferation of artificial intelligence (AI) applications in edge computing environments has led to a diverse array of hardware accelerators, ranging from low-power microcontrollers to high-performance neural processing units. This heterogeneity presents significant challenges for deploying machine learning (ML) models, particularly in balancing model accuracy with resource efficiency. In this paper, we introduce Juggler, a framework designed to optimise ML model serving across heterogeneous edge infrastructures. Juggler employs a knowledge-based routing mechanism encompassing three strategies: task-aware, model-aware, and device-aware to allocate inference workloads based on user-defined multi-criteria optimisation. By profiling device-specific performance characteristics and utilising polynomial regression representations, Juggler effectively manages workload distribution, enhancing real-time inference on resource-constrained hardware.
- 20 June 2026
- Published by Association for Computing Machinery (ACM)
- p. 13-18
Open AccessJuggler: Accuracy v/s Efficiency - Close View on Heterogeneous Embedded Edge ML Serving
The proliferation of artificial intelligence (AI) applications in edge computing environments has led to a diverse array of hardware accelerators, ranging from low-power microcontrollers to high-performance neural processing units. This heterogeneity presents significant challenges for deploying machine learning (ML) models, particularly in balancing model accuracy with resource efficiency. In this paper, we introduce Juggler, a framework designed to optimise ML model serving across heterogeneous edge infrastructures. Juggler employs a knowledge-based routing mechanism encompassing three strategies: task-aware, model-aware, and device-aware to allocate inference workloads based on user-defined multi-criteria optimisation. By profiling device-specific performance characteristics and utilising polynomial regression representations, Juggler effectively manages workload distribution, enhancing real-time inference on resource-constrained hardware.
- 1 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet Computing
- Vol. 30 (3) , 81-88
The Troubleshooter That Never Sleeps: Continually Learning Agentic AI for Fault Debugging in IEC Continuum
- 1 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet Computing
- Vol. 30 (3) , 81-88
The Troubleshooter That Never Sleeps: Continually Learning Agentic AI for Fault Debugging in IEC Continuum
- Open Access
- 31 December 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-6
A Hybrid Monitoring Framework for Anomaly Detection in Smart Home IoT Devices
- 31 December 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-6
Open AccessA Hybrid Monitoring Framework for Anomaly Detection in Smart Home IoT Devices
- 30 September 2025
- Published by Wiley in Concurrency and Computation: Practice and Experience
- Vol. 37 (25-26)
IoTSim ‐Osmosis‐ MARL : Towards Multi‐Agent Reinforcement Learning Osmotic Computing
Internet of Things systems exist in various areas of our everyday life. Data from sensors installed in smart cities and homes is processed in edge and cloud computing centers, providing several benefits that improve our lives. The group of devices might cooperate to fulfill desired goals, trying to preserve their resources and handle the failures of the devices. The paper presents the multi‐agent reinforcement learning (MARL) extension to the osmotic computing simulation framework, enabling direct interactions between IoT devices. The proposed approach allows IoT devices to operate autonomously and cooperatively in dynamic environments, reducing the need for manual intervention and enabling resilient, energy‐efficient sensing coverage. We discuss the multi‐agent sensing coverage problem as one directly applicable to the IoT sensing systems. We identify the challenges posed to the framework and analyze management algorithms for cooperating osmotic agents. In the evaluation, we demonstrate that cooperation between devices enables the self‐autonomous behavior of IoT systems. A case study yields promising results, showing that the adaptation of sensors may allow them to replace each other by lowering energy usage (up to 25% reduction), increasing global coverage (from 74% to 92%), and preserving battery life by dynamically adjusting their sensing range. Finally, the presented framework is a novel contribution that combines MARL environments with IoT systems simulation, enabling future research in this field.
- 30 September 2025
- Published by Wiley in Concurrency and Computation: Practice and Experience
- Vol. 37 (25-26)
IoTSim ‐Osmosis‐ MARL : Towards Multi‐Agent Reinforcement Learning Osmotic Computing
Internet of Things systems exist in various areas of our everyday life. Data from sensors installed in smart cities and homes is processed in edge and cloud computing centers, providing several benefits that improve our lives. The group of devices might cooperate to fulfill desired goals, trying to preserve their resources and handle the failures of the devices. The paper presents the multi‐agent reinforcement learning (MARL) extension to the osmotic computing simulation framework, enabling direct interactions between IoT devices. The proposed approach allows IoT devices to operate autonomously and cooperatively in dynamic environments, reducing the need for manual intervention and enabling resilient, energy‐efficient sensing coverage. We discuss the multi‐agent sensing coverage problem as one directly applicable to the IoT sensing systems. We identify the challenges posed to the framework and analyze management algorithms for cooperating osmotic agents. In the evaluation, we demonstrate that cooperation between devices enables the self‐autonomous behavior of IoT systems. A case study yields promising results, showing that the adaptation of sensors may allow them to replace each other by lowering energy usage (up to 25% reduction), increasing global coverage (from 74% to 92%), and preserving battery life by dynamically adjusting their sensing range. Finally, the presented framework is a novel contribution that combines MARL environments with IoT systems simulation, enabling future research in this field.
- Open Access
- 4 September 2025
- Published by Association for Computing Machinery (ACM) in ACM Computing Surveys
- Vol. 58 (1) , 1-32
Cybersecurity Challenges in the EV Charging Ecosystem
The growing adoption of intelligent Electric Vehicles (EVs) has also created an opportunity for malicious actors to initiate attacks on the EV infrastructure, which can include a number of data exchange protocols across the various entities that are part of the EV charging ecosystem. These protocols possess a range of underlying vulnerabilities that attackers can exploit to disrupt the regular flow of information and energy. While researchers have considered vulnerabilities of particular components within an EV charging ecosystem, there is still a notable gap in vulnerability analysis of charging protocols and the potential threats to these. We investigate threat vectors within the most widely adopted protocols used in EV infrastructure, explore the potential impact of cyberattacks and suggest various mitigation techniques investigated in literature. Potential future research directions are also identified.
- 4 September 2025
- Published by Association for Computing Machinery (ACM) in ACM Computing Surveys
- Vol. 58 (1) , 1-32
Open AccessCybersecurity Challenges in the EV Charging Ecosystem
The growing adoption of intelligent Electric Vehicles (EVs) has also created an opportunity for malicious actors to initiate attacks on the EV infrastructure, which can include a number of data exchange protocols across the various entities that are part of the EV charging ecosystem. These protocols possess a range of underlying vulnerabilities that attackers can exploit to disrupt the regular flow of information and energy. While researchers have considered vulnerabilities of particular components within an EV charging ecosystem, there is still a notable gap in vulnerability analysis of charging protocols and the potential threats to these. We investigate threat vectors within the most widely adopted protocols used in EV infrastructure, explore the potential impact of cyberattacks and suggest various mitigation techniques investigated in literature. Potential future research directions are also identified.
- 29 May 2025
- Published by Elsevier in Future Generation Computer Systems
- Vol. 174
MicroFaaS: Adaptive serverless computing for Internet of Things
- 29 May 2025
- Published by Elsevier in Future Generation Computer Systems
- Vol. 174
MicroFaaS: Adaptive serverless computing for Internet of Things
- Open Access
- 31 March 2025
- Published by Association for Computing Machinery (ACM)
- p. 1612-1616
Automating Data Quality Monitoring for Environmental Air Quality System
- 31 March 2025
- Published by Association for Computing Machinery (ACM)
- p. 1612-1616
Open AccessAutomating Data Quality Monitoring for Environmental Air Quality System