On-device Online Learning and Semantic Management of TinyML Systems
2024 - ACM Transactions on Embedded Computing Systems,Vol. 23, Issue 4
Recent advances in Tiny Machine Learning (TinyML) empower low-footprint embedded devices for real-time on-device Machine Learning (ML). While many acknowledge the potential benefits of TinyML, its practical implementation presents unique challenges.This study aims to bridge the gap between…
CLINT: Controller-Assisted Lightweight In-Band Network Telemetry
2024 - IEEE Access, Vol. 12
The increasing need for high-precision monitoring and network-wide observability has spurred significant interest in In-Band Network Telemetry (INT), which empowers the collection of monitoring data directly from the dataplane, thus, alleviating the various shortcomings of passive monitoring…
A Platform for Time-Sensitive Networking in Converged IoT-Cloud Environments, Demo Track
2024 - 27th IEEE Conference on Innovation in Clouds, Internet and Networks (ICIN 2024)
IoT-enabled applications benefit from various forms of digital twinning, such as the so-called Virtual Objects (VOs). A crucial requirement in the communication among IoT devices and their associated VOs is low latency. In this respect, we showcase a software platform for Time-Sensitive Networking…
Dynamic Schedule Computation for Time-Aware Shaper in Converged IoT-Cloud Environments
2024 - 27th IEEE Conference on Innovation in Clouds, Internet and Networks (ICIN 2024)
Over the last years, we have seen the emergence of (hyper-)distributed applications, which tend to be provisioned over resources that span IoT devices, IoT gateways, and compute nodes from edge computing infrastructures. The deployment and orchestration of such services benefits from types of IoT…
Virtual On-board Unit Migration in a Multi-access Smart-Campus 5G Architecture
2023 IEEE Future Networks World Forum
By embracing the next generation of network architectures, academic institutions are setting the scene to expand the research capabilities, enhance the learning experience of the students, and showcase how smart-city-like solutions can transform traditional educational environments into innovative…
Intent-driven orchestration of serverless applications in the computing continuum
2024 Future Generation Computer Systems Volume 154, May 2024, Pages 72-86
Following the emergence of Internet of Things (IoT) and edge computing technologies and the development of distributed applications with strict Quality of Service (QoS) requirements, a transition is underway from centralized computing models to distributed computing continuum systems. The latter…
A Virtual Object Stack for IoT-Enabled Applications Across the Compute Continuum
2023 International Workshop on the Cloud-Edge-IoT Continuum for Hyper-Distributed Applications and Services (CEICO)
5G and cloud computing are set out to address various business needs, often related to the deployment of IoT-enabled applications across the compute continuum. However, the provision of such hyper-distributed applications is far from trivial, due to various interoperability and convergence aspects…
ROS-based Robotic Applications Orchestration in the Compute Continuum: Challenges and Approaches
2023 International Workshop on the Cloud-Edge-IoT Continuum for Hyper-Distributed Applications and Services (CEICO)
With the adoption of robots growing in several industrial sectors (e.g., logistics, healthcare, agriculture) comes the experience that in "robotic applications" robots are but components of larger distributed systems and, despite their specific requirements and assumptions, should be integrated…
TinyMetaFed: Efficient Federated Meta-Learning for TinyML
2023 ECML PKDD workshop track: Simplification, Compression, Efficiency, and Frugality for Artificial Intelligence (SCEFA) , Torino, Italy
The field of Tiny Machine Learning (TinyML) has made substantial advancements in democratizing machine learning on lowfootprint devices, such as microcontrollers. The prevalence of these miniature devices raises the question of whether aggregating their knowledge can benefit TinyML applications.…
TinyReptile: TinyML with Federated Meta-Learning
2023 International Joint Conference on Neural Networks (IJCNN) , Gold Coast, Australia
Tiny machine learning (TinyML) is a rapidly growing field aiming to democratize machine learning (ML) for resource-constrained microcontrollers (MCUs). Given the pervasiveness of these tiny devices, it is inherent to ask whether TinyML applications can benefit from aggregating their knowledge.…