About
619
Publications
113,395
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
8,515
Citations
Introduction
Piero Castoldi has been Professor in Telecommunications area-leader of the "Networks and Services" area since 2001, and Coordinator of the "Communications" Macro-area since Jan. 2012 at the TeCIP Institute at Scuola Superiore Sant'Anna in Pisa (Italy).
He has been involved in several european projects (STRONGEST, e-photon/ONE, e-photon/ONE+, BONE and has been coordinator of the research activity of several corporate-sponsored projects (with Ericsson and Rete Ferroviaria Italiana).
Publications
Publications (619)
A comprehensive framework for pervasive telemetry and accelerated networking is presented. The framework is suitable for effective integration of optical, packet, and computing resources, leveraging DPU and P4 programmability in support of edge-to-edge continuum.
This paper investigates the main highlights of optical network control evolution, focusing on Software-Defined Networking (SDN), NETCONF/YANG protocols, telemetry techniques, advancements in packet/optical networking, and the integration of Artificial Intelligence (AI) within optical networks.
Power consumption of devices and network functionalities in optical infrastructures is reviewed. Then, possible short-, medium-, and long-term solutions to reduce and make energy consumption scalable are discussed.
OpenSDK is proposed to provide vendor-neutral, micro-service-based control of underlying optical hardware. Disaggregation is then achieved without requiring standard Southbound interfaces from the SDN Controller. Validation is performed enforcing smart operations on IPoWDM white box.
Data Processing Units (DPUs) with embedded GPU have the potential to rev-olutionize optical networks functionalities at the edge. Use cases are presented for optical data monitoring with local AI processing and embedded security.
Swarm devices are becoming increasingly prevalent in next-generation networks because they can efficiently handle large-scale distributed computing tasks. The traditional cloud-based frameworks are not suitable for swarm devices due to their low latency and scalability requirements. Moreover, these frameworks are vulnerable to security threats when...
Space division multiplexing (SDM) is being investigated in support of traffic growth. SDM networks offer multiplexing over space and spectrum dimensions. Probabilistic constellation shaping (PCS) is a solution to optimize spectral efficiency by adapting the constellation of a M-ary quadrature amplitude modulated (QAM) signal to physical layer impai...
In recent years, the adoption of innovative technologies in maritime transport and logistics systems has become a key aspect towards their development and growth, especially due to the complex and heterogeneous nature of the maritime environment. On the other hand, Internet of Things (IoT) solutions are gaining importance in the shipping industry t...
Traditional metro networks are composed of packet switching nodes (i.e., routers) interconnected by optical transport links. In this scenario, the packet and optical domains are clearly separated, using dedicated controllers. Standalone muxponders/transponders will be replaced in optical metro and transport networks by the utilization of hybrid pac...
We propose mode-group (MG) division multiplexing in a MIMO-free configuration to increase the capacity and the flexibility in reduced-complexity spatial division multiplexed metro networks. We take in account innovative few mode fibers (FMFs) supporting a large number of spatial modes, which allow to select suitable combinations of MGs providing li...
In the modern era of artificial intelligence, increasingly sophisticated artificial neural networks (ANNs) are implemented, which pose challenges in terms of execution speed and power consumption. To tackle this problem, recent research on reduced-precision ANNs opened the possibility to exploit analog hardware for neuromorphic acceleration. In thi...
This paper presents the potentials and challenges of programmable packet-optical nodes encompassing coherent pluggable modules applied in the context of optical metro networks. Two innovative applications of P4-based data plane programmability are then presented. The first targets the monitoring and processing of optical telemetry data/metadata dir...
Nowadays, the fiber spectrum is only partially exploited, i.e., mainly in the C-band and more recently in the C $+$ + L-band, where the fiber attenuation profile experiences the minimum. Thus, fiber communications technology—amplifiers, switching, transceivers, etc.—and networking solutions are mature for those spectrum bands. However, the continuo...
Autonomic networking and monitoring will drive the evolution of next generation software defined networking (SDN) optical networks towards the zero touch networking paradigm. Optical telemetry services will play a key role to enable advanced network awareness at device and component granularity. Optical disaggregation is pushing the adoption of ope...
Service chaining is becoming one of the most considered service deployment frameworks in the context of Network Function Virtualization (NFV) in edge and data center environments, conveniently supported by automatic connectivity configurations offered by Software Defined Networking (SDN). Current research on the topic is focusing on how to guarante...
Edge Computing, along with Software Defined Networking and Network Function Virtualization, are causing network infrastructures to become as distributed clouds extended to the edge with services provided as dynamically established sequences of virtualized functions (i.e., dynamic service chains) thereby elastically addressing different processing r...
In the modern era of artificial intelligence, increasingly sophisticated artificial neural networks (ANNs) are implemented, which pose challenges in terms of execution speed and power consumption. To tackle this problem, recent research on reduced-precision ANNs opened the possibility to exploit analog hardware for neuromorphic acceleration. In thi...
This paper presents potentials and challenges of disaggregated metro-edge networking based on open packet-optical nodes encompassing coherent pluggable modules, SONiC open operating system, and P4-based packet switching programmability.
Software Defined Networking represents a mature technology for the control of optical networks, though all open controller implementations present in the literature still lack the adequate level of maturity and completeness to be considered for (pre)-production network deployments. This work aims at experimenting on, assessing and discussing the us...
Exploiting bands beyond C+L can effectively upgrade network capacity, but Stimulated Raman Scattering (SRS) affects wideband-transmission, potentially degrading active channels. Upgrades exploiting E- and S-band are compared in terms of capacity and number of reconfigurations.
Space division multiplexing (SDM) – such as based on multi-mode fibers – is under investigation to support the growing capacity demands. In optical networks, the estimation of quality of transmission (QoT) is a fundamental control task to identify the most suitable transmission parameters (e.g., modulation format) to meet proper spectral efficiency...
Few Mode Fibers (FMF) and Space Division Multiplexing (SDM) are an attractive solution to offer high capacity in optical networks. Although transmission along FMF presents several issues mainly due to the cross-talk among modes, the use of Multiple Input Multiple Output (MIMO) coherent receivers permits to limit the impact of such physical impairme...
In the last twenty years, optical networks have witnessed recurrent changes in their management and control architecture. In this paper, we present an historical timeline and a future perspective of the evolution of optical network management and control deployed for Wavelength Switched Optical Networks (WSON), Elastic Optical Networks (EON) and (m...
The first implementation of the OneM2M IoT platform used to control optical networks is presented. The platform provides remarkable scalability performance particularly as a function of the number of monitored devices.
This paper proposes a new paradigm for control plane in Time Sensitive Networks (TSN). An SDN controller proactively instructs network elements on the reconfigurations to perform locally if some specific events occur (e.g., failures, performance degradations). Instructions are given in the form of Finite State Machines (FSMs), which store informati...
The new radio technologies (i.e. 5G and beyond) will allow a new generation of innovative services operated by vertical industries (e.g. robotic cloud, autonomous vehicles, etc.) with more stringent QoS requirements, especially in terms of end-to-end latency. Other technological changes, such as Network Function Virtualization (NFV) and Software-De...
Optical network disaggregation is attracting significant consensus to avoid vendor lock-in solutions. However, the presence of network controllers, nodes, and hardware/software components potentially provided by different entities and manufacturers may lead to remarkable responsibility issues in case of service level degradation. We propose the use...
In the modern era of artificial intelligence, increasingly sophisticated artificial neural networks (ANNs) are implemented, which pose challenges in terms of execution speed and power consumption. To tackle this problem, recent research on reduced-precision ANNs opened the possibility to exploit analog hardware for neuromorphic acceleration. In thi...
Reconfigurable linear optical processors can be used to perform linear transformations and are instrumental in effectively computing matrix–vector multiplications required in each neural network layer. In this paper, we characterize and compare two thermally tuned photonic integrated processors realized in silicon-on-insulator and silicon nitride p...
Telemetry data acquisition is becoming crucial for efficient detection and timely reaction in the case of network status changes, such as failures. Streaming telemetry data to many collectors might be hindered by scalability issues, causing delay in localization and detection procedures. Providing efficient mechanisms for managing the massive telem...
Effective telemetry streaming for disaggregated whitebox is discussed, focusing on implementation issues. A novel gRPC peer-to-peer telemetry for next-generation AI-equipped xPonders is proposed and evaluated in a disaggregated testbed exploiting OpenConfig and OpenROADM models.
A programmable P4 node performing neural network acceleration is demonstrated. The node implements both traffic feature extraction and neural networking computations. Cyber security use case targeting DDoS attack detection is deployed. © 2021 The Author(s) 1. Introduction Bringing in-network Artificial Intelligence (AI) is hot topic in the developm...
This demo shows a 5G X-haul testbed enhanced
with P4 switches implementing the offloading of the User
Plane Function module. The P4 code includes GTP protocol
encapsulation/decapsulation function, fully configurable N3-N6-
N9 steering, and advanced online monitoring of the experienced
latency metadata.
5G Networks revolution will be enabled by deep integration of Software Defined Networking (SDN) and Network Function Virtualization (NFV) to support multi-tenancy, per-user and per-application quality of service and experience. However, full softwarization and current SDN platforms may not be able to sustain the complexity and the heterogeneity of...
Optical interconnection networks promise to overcome the limitations of current electronic switching fabrics, enabling higher throughput, lower latency, and lower power consumption. Multi-plane architectures, based on multiple optical switching domains (e.g., space, time, wavelength, orbital angular momentum), are gaining research attention because...
Currently, most of the Emergency Service interventions involve as operators on board of ambulances only nurses and volunteers, without the presence of doctors. In this work REC-VISIO 118, a new system of remote medical tutoring to the operators involved in emergency interventions, is illustrated. The system is composed by a subsystem worn by the op...
We present a reduced-precision integrated photonic electronic multiply-accumulate (MAC) neuron with ADC-embedded nonlinearity. The proposed device trades off speed with resolution, outperforming both analog and digital electronic solutions in terms of speed and energy consumption. © 2021 The Author(s) 1. Introduction In recent years, deep neural ne...
This demo shows how crowd management can benefit from 5G connectivity and accelerated edge cloud AI-based computation to achieve low latency and scalability with respect to a mobile device-based physical distancing computation.
Effective control of pluggable transceivers in SONiC-based packet-optical nodes is demonstrated. A workflow for multi-layer recovery upon soft failure detection is validated, showing no traffic disruption and fast node-driven coordination between packet and optical operations.
The characterization of a broadband Si 3 N 4 integrated linear optical processor operating in the C-band is reported. The impact of losses on the processor accuracy is discussed towards the photonic implementation of state-of-the-art neural networks.
This paper presents Mode Group Division Multiplexing (MGDM), discussing transmission aspects, node architecture, and provisioning.
This paper provides an overview of state of art and challenges for open packet-optical node solutions based on SONiC open operating system, coherent pluggable modules and P4-based packet switching programmability for edge networking.
Traffic monitoring through in-band telemetry is extended up to the User Equipment (UE), providing accurate e2e latency measurement. The UE becomes aware of its experienced service performance, enabling autonomous operations for faster automatic source-based Edge-Cloud steering.
Network services in 5G will be rolled out as pools of virtual network functions (VNFs) exploiting the advantages of both software-defined networking and network function virtualization. In this context, 5G network services are envisaged as ordered sequences of VNFs resulting in the so-called VNF Forwarding Graphs (VNFFGs). Such VNFs can be allocate...
Fifth generation softwarized network systems will make it possible to flexibly partition the network infrastructure into logically independent network slices, hosting end-to-end network services able to dynamically meet the diverse requirements of vertical industries. However, the high dynamicity of NFV-related operations and the interdependence of...