Title: On Optimal Offloading of DNNs from IoTs to Cloud
Speaker: Prof. Jie Wu
Abstract: As Deep Neural Networks (DNNs) have been widely used in various applications, including computer vision on image segmentation and recognition, it is important to reduce the makespan of DNN computation, especially when running on IoT devices. Offloading is a viable solution that offloads computation from a slow IoT device to a fast, but remote server in cloud. As DNN computation consists of a multiple-stage processing pipeline, it is critical to decide on what stage should offloading occur to minimize the makespan. Our observations show that the local computation time on a mobile device follows a linear increasing function, while the offloading time on a mobile device is monotonic decreasing and follows a convex curve as more DNN layers are computed in the mobile device. Based on this observation, we first study the optimal partition and scheduling for one line-structure DNN. Then, we extend the result to multiple line-structure DNNs. Heuristic results for general-structure DNNs, represented by Directed Acyclic Graphs (DAGs), are also discussed based on a path-based scheduling policy. Our proposed solutions are validated via real system implementation.
Bio: Jie Wu is Laura H. Carnell Professor at Temple University and the Director of the Center for Networked Computing (CNC). He served as Chair of the Department of Computer and Information Sciences from the summer of 2009 to the summer of 2016 and Associate Vice Provost for International Affairs from the fall of 2015 to the summer of 2017. Prior to joining Temple University, he was a program director at the National Science Foundation and was a distinguished professor at Florida Atlantic University. His current research interests include mobile computing and wireless networks, routing protocols, network trust and security, distributed algorithms, applied machine learning, and cloud computing. Dr. Wu regularly published in scholarly journals, conference proceedings, and books. He serves on several editorial boards, including IEEE Transactions on Service Computing, IEEE/ACM Transactions on Networking, and Journal of Computer Science and Technology. Dr. Wu is/was general chair/co-chair for IEEE IPDPS’23, ACM MobiHoc’23, and IEEE CCGrid 2024 as well as program chair/cochair for IEEE INFOCOM’11, CCF CNCC’13, and ICCCN’20. He was an IEEE Computer Society Distinguished Visitor, ACM Distinguished Speaker, and chair for the IEEE Technical Committee on Distributed Processing (TCDP). Dr. Wu is a Fellow of the AAAS and a Fellow of the IEEE. He is a Member of the Academia Europaea (MAE).
Title: Edge Sensing as a Service: Unlocking IoT Virtualization and Decentralized Data Markets
Speaker: Prof. Marco Di Felice
Abstract: Abstracting physical hardware resources to create virtual execution environments has been one of the most impactful trends in software engineering over the past two decades. While virtualization technologies have paved the way for scalable and large-scale Internet of Things (IoT) deployments, one segment of IoT infrastructure remains largely unaffected: extreme edge computing. This layer, composed of IoT end nodes such as microcontrollers, is typically managed as a semi-static software environment with limited multitenancy capabilities.
In this talk, we investigate a paradigm shift through Edge Sensing as a Service (ESaS)—an approach that abstracts IoT computing and sensing resources to enable the development of secure, efficient, and customizable applications while unlocking new monetization opportunities. We explore the deployment of ESaS in two complementary directions.
First, we discuss how to extend containerization and software orchestration to IoT end devices, allowing multi-tenant applications to securely share access to sensing peripherals. This requires addressing critical research challenges in embedded system security, sensing resource virtualization, and container orchestration.
Second, we explore how ESaS can enable a global IoT data marketplace, allowing users to seamlessly access high-quality data from IoT end nodes worldwide while providing monetization opportunities for device owners. To achieve this, we introduce ZONIA, a decentralized IoT market architecture that leverages a distributed oracle layer built on top of smart contracts and powered by a global network of IoT end devices. We discuss the key research challenges behind ZONIA, including automatic discovery, interoperability mechanisms, and reputation-based algorithms to ensure trustworthy data sources selection.
Bio: Marco Di Felice obtained a Ph.D. in Computer Science from the University of Bologna in 2008. He was a visiting researcher at the Georgia Institute of Technology and Northeastern University, USA. Since 2022, he has been a Full Professor at the Department of Computer Science and Engineering at the University of Bologna, where he co-founded and coordinates the IoT Prism Laboratory, dedicated to research and teaching in pervasive and mobile systems.
His research interests include the Internet of Things (IoT) and Artificial Intelligence of Things (AIoT), pervasive, autonomous, and context-aware systems, edge-cloud computing and edge AI techniques. He has authored over 180 publications in international journals and conference proceedings on the analysis, modeling, and performance evaluation of pervasive and mobile systems, receiving three Best Paper Awards. He has served on the editorial boards of several journals, including Ad Hoc Networks, Computer Networks, and the IEEE Internet of Things Journal.