Pr. Pierre Boulet
Pierre Boulet received a PhD of computer science in 1996 from the École Normale Supérieure de Lyon, France. He is currently professor of computer science at the University of Lille, France.
His interests range from parallelism, compilation, embedded system co-design to model driven engineering and synchronous languages. He is currently investigating how to program time and energy aware embedded applications on post-Moore architectures, and help design neuromorphic accelerators. He initiated the research on neuromorphic computing at CRIStAL and IRCICA with colleagues from nanoelectronics (IEMN lab) in 2012.
He was deputy director of LIFL then CRIStAL labs from 2008 to 2017. He was vice-president for digital transformation of the University of Lille between 2018 and 2021 and has been vice-president for digital infrastructures since 2022. He is a member of the HiPEAC EU support action, and senior member of the IEEE and ACM professional societies.
Designing neuromorphic architectures: towards an ultra low power AI
In the "bio-inspired information processing" project of the IRCICA interdisciplinary institute, we tackle the scientific challenges of the emerging neuromorphic architectures. These computer architectures mimic the brain by handling the information as spike trains and by processing this information with spiking neural networks. They have a strong potential for ultra low power artificial intelligence. Based on our last 10 years of research, we will present the state-of-the-art of these architectures, the applications we focus on, and the scientific hot topics.
Pr. Maryline Chetto
Maryline Chetto is currently a full professor in computer engineering with Nantes Université, France and researcher with CNRS. She received the degree of Docteur de 3ème cycle in control engineering and the degree of Habilitée à Diriger des Recherches in Computer Science from the University of Nantes, France, in 1984 and 1993, respectively. From 1984 to 1985, she held the position of Assistant professor of Computer Science at the University of Rennes, while her research was with the Institut de Recherche en Informatique et Systèmes Aléatoires (IRISA), Rennes. In 1986, she returned to Nantes and has been from 2002 a full professor with the University of Nantes. She is conducting her research at Laboratoire des Sciences du Numérique de Nantes (LS2N, UMR CNRS n° 6004) in the Real Time System group.
Her research has been focused on development and formal validation of solutions regarding Scheduling, Fault-tolerance and Dynamic Power Management in real time embedded applications. Her current research is specifically targeting real-time scheduling issues in energy neutral devices. She has more than 150 papers published in international journals and conferences. She was the editor of the books Real-time Systems Scheduling (Elsevier, 2014) with volume 1 Fundamentals and Volume 2 Focusses. She was the co-author of the book Energy Autonomy of real-time systems (Elsevier, 2016). She was general chair of the 2020 IEEE International Conference on Green Computing and Communications (GreenCom-2020). Since 2011, she was elected member of the French National Council for Universities.
Challenges in Real-time Scheduling for Energy Harvesting Embedded Systems
The recent and rapid development of the Internet of Things (IoT) leads to the need for embedded devices that comprise one or more resource-constrained computing elements and sensors. Energy Harvesting (EH) technology provides the ability for any low-power embedded device to operate thanks to the generation of electricity from the energy available in the device’s immediate surroundings such as light or motion. Energy neutrality is the central requirement of autonomous real-time computing systems that should be designed to correctly function for long times with no possible manual intervention to charge or replace batteries. Unfortunately, most of environmental energy sources are fluctuating and not controllable. It means that a stable power supply cannot be relied upon that makes challenging the issue of compliance with hard real-time constraints. Specific power management and scheduling solutions have to be conceived in order to prevent energy starvation and guarantee real-time responsiveness. Task scheduling should take into account not only the timing parameters of the deadline constrained tasks such as worst-case execution times but also energy consumptions, profile of the energy source and capacity of the energy storage unit. The classical greedy scheduler Earliest Deadline First (EDF) or Rate Monotonic (RM) used in battery powered devices should be revisited for this novel operational context.
This keynote discusses the state of the art as well as our findings in real-time scheduling and dynamic processor management for autonomous embedded systems.
Pr. Fulvio Risso
Fulvio Risso is Associate Professor with the Department of Control and Computer Engineering of Politecnico di Torino, Italy. He received the M.Sc. degree in Computer Engineering in 1995 and the Ph.D. in Computer and System Engineering in 2000, both from Politecnico di Torino.
He has been visiting student at University College London (UK), visiting Faculty at Cisco Systems (San Jose, CA) and Narus (Sunnivale, CA); adjunct professor at University of Illinois at Chicago (UIC), and Turin Polytechnic in Tashkent (TTPU), Uzbekistan.
His recent research interests include high speed processing of network traffic, network functions virtualization, and edge/fog/cloud computing. He co-authored more than 140 articles published in the most renowned international journals, magazines, and conference proceedings. He served as Technical Program Co-Chair for IEEE NetSoft 2022, and guest editor for the IEEE Communication Magazine special issue on fog computing in 2017.
He is very active in open-source software projects, starting from WinPcap, for 15+ years the de-facto standard library for network packet capture in Windows, NetBee, a novel packet processing library including both a hardware abstraction (NetVM) and a protocol description language (NetPDL), Polycube, an efficient e-BPF-based framework for Network Functions Virtualization, and Liqo, a multi-cluster and multi-domain cloud-to-edge solution for Kubernetes.
Creating an Edge-to-Cloud Computing Continuum: Status and Perspective
The creation of a computing (and storage) continuum that spans across multiple infrastructures has been recently proposed with the aim at breaking the barriers of current cloud computing silos. While the potential of this vision is clear, such as possible energy efficiency improvements, application transparency, better quality of service, the problems in realizing this vision are challenging. In fact, the necessity (a) to consider WAN latencies, network bottlenecks and possible network outages, (b) to support different administrative boundaries (hence, security), (c) to define novel orchestration algorithms are only a first portion of the ingredients required to create this continuum. Even more important are the necessity to achieve an economic sustainability and the necessity to migrate applications to the new paradigm, which are often not considered by the current research efforts.
Starting from the experience achieved in the open-source liqo.io project, this talk will introduce the potentials and the challenges in the creation of the edge-to-cloud computing continuum, and it will present what are the major blocking points along our way.
Pr. Wahab Hamou-Lhadj
Pr. Wahab Hamou-Lhadj is a Professor in the Department of Electrical and Computer Engineering at the Gina Cody School of Engineering and Computer Science, Concordia University, Montreal, QC, Canada. His research interests are in software engineering, software tracing and logging, system observability, model-driven engineering, and applications of AI to computing systems. He has been the principal investigator for several projects with various organizations including Ericsson, Ericsson Global AI Accelerator, CAE, Ubisoft, Marivent, Opal-RT, and Defence Canada. Several of the original tools that were developed in his lab (e.g., TotalADS and CommitAssistant) have been successfully transferred to the industry and are currently used by thousands of developers. Pr. Hamou-Lhadj served on the organization and program committees of major conferences in software engineering such as ICSE, SANER, ICPC, ICSME, MODELS to name a few. In 2020, he was the General Co-chair of the 12th ACM System Analysis and Modelling Conference (SAM). Pr. Hamou-Lhadj received his PhD from the University of Ottawa, ON, Canada. He is a senior member of IEEE, IEEE Computer Society, and a member of ACM. He is also a frequent contributor to the Object Management Group (OMG) certification programs, OCUP 2 and OCEB 2.
Observability of Software Computing Systems: Challenges and Opportunities
Modern computing systems expand on the challenges of traditional monolithic applications by relying on a combination of servers, embedded and sensory devices, diverse architectural models, and various communication mechanisms to drive new ways of creating value and stimulating growth in diverse sectors of modern society. The fragmented and distributed nature of these systems call for advanced system analysis and fault diagnosis methods-timely detection and prevention of crashes and anomalies are of paramount importance. In this talk, we will start by discussing the concept of system observability, which we present as the umbrella field for the current research that aim to gain insight into runtime data (traces, logs, profiling metrics, etc.). we will then dive into the research opportunities and challenges in system observability using log analytics with an emphasis on software and hardware system analysis.
Pr. Smail Niar
Pr. Smail Niar, Université Polytechnique Hauts-de-France (UPHF) & CNRS, France, received his PhD in computer Engineering from the University of Lille (France) in 1990. Since then, he has been professor at UPHF where he is director of the computer science department at the “Laboratory of Automation, Mechanical and Computer Engineering”, a joint research unit between CNRS and UPHF. S.Niar co-supervises the «Intelligent infrastructures & vehicles » task within the International Campus on Safety and Inter-modality in Transportation (“Campus International sur la Sécurité et l'Intermodalité dans les Transports” CISIT). Prof. S.Niar is member of the editorial board at the “Embedded Hardware Design: Microprocessors and Microsystems” (Elsevier) journal. He is member of the European Network of Excellence on “HIgh Performance and Embedded Architectures and Compilation” (HIPEAC), EuroMicro society and IEEE senior member. His research interests are in heterogeneous multi-processor system-on-chip (MPSoC) architectures in intelligent transportation systems, power/energy consumption optimization, dynamically reconfigurable embedded systems (FPGA) and reliability issues for embedded systems.
Optimizing Deep Learning Application for Edge Computing
Deep learning (DL) models such as convolutional neural networks (CNN) are being deployed to solve various computer vision and natural language processing tasks at the edge. It is a challenge to find the right DL architecture that simultaneously meets the accuracy, power and performance budgets of such resource-constrained devices. Hardware-aware Neural Architecture Search (HW-NAS) has recently gained steam by automating the design of efficient DL models for a variety of target hardware platform. However, such algorithms require excessive computational resources. Thousands of GPU days are required to evaluate and explore modern DL architecture search space. In this talk I will present state-of-the-art approaches that are based on two components: a) Surrogate models to predict quickly architecture accuracy and hardware performances to speed up HW-NAS, b) Efficient search algorithm that explores only promising hardware and software regions of the search space.