Banner
Home      Log In      Contacts      FAQs      INSTICC Portal
 

Keynote Lectures

"Living in the Edge, Sailing Through the Cloud”: Orchestrating Applications in the Edge to Cloud Computing Continuum
Tamas Kiss, University of Westminster, United Kingdom, United Kingdom

Integrating Machine Learning and Multi-Agent Systems for Fully Enabling Device-Edge-Cloud Continuum in Complex IoT Worlds
Giancarlo Fortino, University Calabria, Italy, Italy

 

"Living in the Edge, Sailing Through the Cloud”: Orchestrating Applications in the Edge to Cloud Computing Continuum

Tamas Kiss
University of Westminster, United Kingdom
United Kingdom
https://www.westminster.ac.uk/about-us/our-people/directory/kiss-tamas
 

Brief Bio
Tamas Kiss is a Professor of Distributed Computing at the School of Computer Science and Engineering and Director of the Research Centre for Parallel Computing at the University of Westminster, London, UK. He also servrs as Co-Editor in Chief at the Journal of Grid Computing, published by Springer Nature. He holds a PhD in Distributed Computing, and MSc Degrees in Mathematics and Computer Science, and Electrical Engineering. He has attracted over £60 Million research funding and have been leading national and European research projects related to enterprise and scientific applications of cloud computing technologies. He has been involved in more than 20 European research projects as principal investigator or co-investigator. He has coordinated three EU: CloudSME (Cloud-based Simulation Platform for Manufacturing and Engineering) that developed a Cloud-based simulation solution for manufacturing and engineering SMEs; COLA (Cloud Orchestration at the Level of Application) that investigated how a generic and pluggable framework that supports the optimal and secure deployment and run-time orchestration of cloud applications can be created; and ASCLEPIOS (Advanced Secure Cloud Encrypted Platform for Internationally Orchestrated Solutions in Healthcare) that focused on secure cloud-enabled applications in healthcare. Currently he is involved in five EU projects, including DIGITbrain - Digital twins bringing agility and innovation to manufacturing SMEs, by empowering a network of DIHs with an integrated digital platform that enables Manufacturing as a Service, PITHIA-NRF - Plasmasphere Ionosphere Thermosphere Integrated Research Environment and Access services: a Network of Research Facilities, HARPOCRATES - Federated Data Sharing and Analysis for Social Utility, CO-VERSATILE - Adaptive and resilient production and supply chain methods and solutions for urgent need of vital medical supplies and equipment, and ARCAFF - Active Region Classification and Flare Forecasting. The leading recent research outputs of his team, in collaboration with several other groups, have been the MiCADO cloud orchestrator solution that enables the automated deployment and run-time management of microservices-based applications in heterogeneous cloud infrastructures, and the CloudSME and CloudiFacturing platforms that support manufacturing and engineering companies to utilise cloud-based high-performance computing services to run simulation and optimisation applications. These solutions have already been utilised by close to 100 companies and generated significant economic impact. 


Abstract
With the intensively increasing utilisation of IoT devices, large amounts of data are collected and needs to be processed close to its sources or transferred to larger and centralised cloud computing resources. Such application scenarios are common, for example, in manufacturing where smart sensors attached to manufacturing machines are collecting valuable information, in smart cities where such IoT devices are present in the streets or attached to vehicles, or in healthcare where wearable devices by patients can collect important information related to their behaviour and bio-signals. The collected large amount of data then needs to be processed, typically by big data analytics or artificial intelligence algorithms. A key question in such cases is where to process such information. Local processing using edge or fog computing nodes closer to the data sources can significantly reduce latency. On the other hand, such computational resources are more volatile and have limitations in storage and computational capacity. In comparison, clouds offer practically unlimited storage and computational capacity, but they also increase latency. A key question is how to distribute such loads between the various layers of the computing infrastructure, and how to deploy and manage such complex, typically microservices-based applications, spanning the edge to cloud computing continuum. In this keynote an edge to cloud application-level orchestrator, called MiCADO, will be presented that was developed within several EU funded projects and that is utilised in multiple sectors, including manufacturing, healthcare, public services and research. Besides outlining the evolution and architecture of the solution, it will also be widely illustrated how the orchestrator serves a wide range of use cases from the above-mentioned domains. Additionally, a vision of a generic architecture towards a holistic approach for edge to cloud application-level orchestration will also be presented. 



 

 

Integrating Machine Learning and Multi-Agent Systems for Fully Enabling Device-Edge-Cloud Continuum in Complex IoT Worlds

Giancarlo Fortino
University Calabria, Italy
Italy
 

Brief Bio
Giancarlo Fortino (IEEE Fellow 2022) is Full Professor of Computer Engineering at the Dept of Informatics, Modeling, Electronics, and Systems of the University of Calabria (Unical), Italy. He received a PhD in Computer Engineering from Unical in 2000. He is also a distinguished professor at Wuhan University of Technology (China), a high-end expert at Huazhong University of Science and Technology (China), a senior research fellow at the Italian ICAR-CNR Institute, CAS PIFI Group international fellow at SIAT (Shenzhen), and Distinguished Lecturer for IEEE Sensors Council, SMC society, and IoT TC. He was also a visiting researcher at ICSI, Berkeley (USA), in 1997 and 1999, and a visiting professor at Queensland University of Technology in 2009. At Unical, he is the chair of the PhD School in ICT, the director of the SPEME lab and of the Radiomics lab, and the director of the Postgraduate Master course in AI-driven Radiomics, as well as co-chair of Joint labs on IoT technologies established between Unical and the WUT, SMU, and HZAU Chinese universities, and the AI-driven Robotics Lab funded with the J.C. Bose University of Science and Technology, YMCA. Fortino is also the scientific responsible of the Digital Health group of the Italian CINI National Laboratory at Unical. He is a Highly Cited Researcher 2020-2025 in Computer Science by Clarivate (the only Italian professor ranked). He had 25+ highly cited papers in WoS, and an h-index of 88 with 33000+ citations in Google Scholar. His research interests include wearable computing systems, e-Health, Internet of Things, and agent-based computing. He is the author of 750+ papers in international journals, conferences, and books. He is (founding) series editor of IEEE Press Book Series on Human-Machine Systems and EiC of Springer Internet of Things series and AE of premier int'l journals such as IEEE TASE (senior editor), IEEE TAFFC-CS, IEEE THMS, IEEE T-AI, IEEE SJ, IEEE JBHI, IEEE OJEMB, IEEE OJCS, Information Fusion, EAAI, etc. He chaired many international workshops and conferences (130+), was involved in a huge number of international conferences/workshops (800+) as an IPC member, is/was a guest editor of many special issues (80+). He is cofounder and CEO of SenSysCal S.r.l., a Unical spinoff focused on innovative IoT systems, and, recently, cofounder and vice-CEO of the spin-off Bigtech S.r.l., focused on big data, AI, and IoT technologies. Fortino is the VP of Cybernetics (term 2026-2027) of the IEEE SMCS, member of the IEEE SMCS ExCom, and former chair of the IEEE SMCS Italian Chapter.


Abstract
Recently the device-edge-cloud paradigm is gaining momentum due to the benefits it could provide for the development of highly effective, efficient, and complex IoT ecosystems of diversified scale. However, there are many issues related to unsupervised control aspects that need to be addressed in order to fully realize the approach and make it fully operative in real complex environments. In order to address such issues, in this talk, we propose an holistic integration of machine learning and multi-agent systems to create a data-driven control architecture capable to autonomically monitor and control the device-edge-cloud continuum. This objective is being developed in the context of the Horizon Europe project named MLSysOps (https://mlsysops.eu/). Some use cases will be proposed to elucidate our current findings.



footer