We propose a novel two-level hierarchical auction model for 5G RAN network slicing that enables financially-aware resource allocation among Mobile Network Operators (MNOs), Mobile Virtual Network Operators (MVNOs), and end users. Compared to prior models that primarily emphasize technical resource utilization, our approach integrates a realistic financial cost framework, including electricity consumption and resale bidding fees into the optimization process. Each level solves a Winner Determination Problem (WDP) using Integer Linear Programming (ILP) and scalable heuristic algorithms. Vickrey–Clarke–Groves (VCG)-based pricing is adopted to ensure incentive compatibility and procedural fairness among bidders. The optimization objective is to maximize the social welfare, defined in this work as the total valuation of accepted requests minus the electricity cost incurred by CU–DU placement. Power consumption is therefore internalized as a cost component in the objective function rather than treated as an independent optimization target. Through extensive simulations over different network topologies and dynamic traffic conditions, we show that the proposed approach achieves higher social welfare compared to baseline methods, while maintaining scalability and satisfying the latency and capacity constraints of heterogeneous network slices.

Reinforcement Learning (RL) has emerged as a powerful tool for automating complex network management tasks, yet its lack of transparency and black-box nature hinder trust and adoption in operational environments. In this work, we focus on explaining the behavior of an $\mathbf{R L}$ agent applied to the problem of network slice admission control. We present a framework that integrates three key components: a Deep Reinforcement Learning (DRL) agent for admission control, an Integer Linear Programming (ILP) model for network slice embedding, and an explanation module for interpreting the DRL agent’s policies, namely Shapley Value Explainable Reinforcement Learning (SVERL). Our analysis aims gives particular attention to cases where the RL agent rejects admitting a network slice request despite sufficient network capacity to provision it, and investigates whether explanations can be used to verify and validate the agent’s behavior prior to deployment approval. Experimental results reveal that the agent’s decisions are primarily influenced by substrate network conditions such as congestion, rather than by the intrinsic characteristics of slice requests. While this conservative policy prevents overload, it also leads to overly cautious rejections. Importantly, the proposed explanation framework provides operators with actionable insights to scrutinize, validate, and refine RL-driven policies before operational deployment.

This demonstration showcases the autonomous creation of optical lightpaths across two geographically optical network testbeds, using TeraFlowSDN (TFS) as a high-level intent-based orchestrator for optical service provisioning. Unlike existing approaches that create lightpaths independently within a single domain, our system highlights a hierarchical control model in which a centralized TFS instance coordinates two heterogeneous domain controllers: a vendor-specific controller at Politecnico di Milano (Italy) and a local TFS instance at Sant’Anna School of Advanced Studies (Italy). Southbound adapters enable the translation of high-level service intents into device-specific configurations, making it possible to integrate different controllers and vendors without modifying the underlying infrastructure. The live demo demonstrates automated provisioning of optical lightpaths triggered via a user-friendly graphical interface. The process includes endpoint discovery, transceiver selection, and lightpath establishment, all performed autonomously across multiple domains to support a video streaming service. This work demonstrates the novelty of hierarchical cross-domain orchestration, showing how TFS can unify multivendor environments under a single platform with minimal configuration overhead. This lays the groundwork for future developments in automated service provisioning, closed-loop control, and scalable cross-domain networking.

Customer Premises Equipment (CPE) plays a fundamental role in enabling Software-Defined Wide Area Networks (SD-WANs) to support emerging applications such as Metaverse environments. These environments demand agile, secure, and high-performance network infrastructure to ensure seamless connectivity, low latency, and dynamic resource allocation. As the Metaverse continues to evolve, the need for flexible, efficient, and open CPE solutions becomes increasingly important. This paper presents an open-source SD-WAN CPE software designed to enhance network agility, scalability, and performance, specifically addressing the stringent requirements of next-generation Internet applications. The proposed solution supports deployment on both virtual machines and bare-metal systems, enabling seamless integration into diverse infrastructures. To achieve high-speed packet processing, it leverages the Data Plane Development Kit (DPDK), ensuring low-latency and high-throughput communication. Furthermore, it incorporates different overlay tunneling mechanisms, supporting both site-to-site and hub-and-spoke topologies via WireGuard tunneling technology for secure and efficient connectivity. A key feature of this implementation is its rapid failover capability, powered by Bidirectional Forwarding Detection (BFD), alongside real-time performance monitoring, enabling proactive network optimization and fault tolerance. By combining these advanced capabilities with an open-source framework, this SD-WAN CPE software provides a robust and cost-effective solution for next-generation network deployments.

The growing reliance on digital connectivity has made Internet Service Provider (ISP) networks a critical component of modern society, yet they remain a prime target for cyber threats. In recent years, cyberattacks against ISPs have increased in scale and sophistication, posing severe risks to national security, economic stability, and user privacy. The advent of in-network computing and programmable data plane presents a paradigm shift in network security, offering the flexibility to define, modify, and optimize packet processing logic dynamically. Among these advancements, the P4 programming language plays a crucial role, allowing network operators to implement fine-grained traffic monitoring directly within network devices. By leveraging in-network computation, P4 facilitates real-time anomaly detection, making it a powerful tool for mitigating Distributed Denial of Service (DDoS) attacks. However, orchestrating security functions across a distributed network of P4 switches remains a challenge, requiring an efficient and scalable deployment framework.In this paper, we present an open-source orchestration platform for managing and deploying P4-based security programs to enable real-time DDoS detection. Our solution leverages dynamic programmability to enhance network security. By integrating a novel queue monitoring mechanism directly into the data plane, our approach enables the collection of fine-grained network performance metrics in real-time, facilitating faster and more precise attack detection and mitigation. The proposed framework is highly scalable and adaptable, strengthening ISP networks against evolving cyber threats.

Network slicing is a key 5G technology that enables multiple virtual networks to share physical infrastructure, optimizing flexibility and resource allocation. This involves Mobile Network Operators (MNO), Mobile Virtual Network Operators (MVNOs), and end users, where MNO leases network slices to MVNOs, and then provides customized services. This work considers end-to-end network slicing with a focus on fair sharing and financial-related power efficiency, modeled as a twolevel hierarchical combinatorial auction. At the upper level, an MNO auctions slices to competing MVNOs, while at the lower level, MVNOs allocate resources to end users through their own auctions. Dynamic user requests add complexity to the process. Our model optimizes resource allocation and revenue generation using a pair-bid mechanism and Vickrey-Clarke- Groves (VCG) pricing. The pair-bid approach enhances competition and efficiency, while VCG ensures truthful bidding based on marginal system impact. Simulations validate the model’s effectiveness in resource distribution and financial performance, showing a $\mathbf{1 2. 5} \boldsymbol{\%}$ revenue improvement over the baseline.

Integrating advanced Machine Learning (ML) techniques, such as Federated Learning (FL), with emerging paradigms like Multi-access Edge Computing (MEC) and Software-Defined Wide Area Network (SD-WAN) has gained significant attention. In this demo, we present a simulator designed to optimize resource allocation for FL in MEC SD-WAN environments, taking into account network capacity constraints. We demonstrate in real-time how resource allocation is implemented for FL algorithms and visually analyze the impact of different constraints on performance. Additionally, we introduce a heuristic algorithm for Distributed Federated Learning (DFL) that selects the optimal global aggregator based on resource utilization. We evaluate the proposed heuristic within our simulation framework and compare its performance against Centralized Federated Learning (CFL) and traditional ML techniques. Our results demonstrate how the proposed simulator effectively integrates network and resource allocation, providing valuable insights into the interplay between FL and MEC SD-WAN infrastructure.

Next-generation networks rely on the network softwarization paradigm to enable faster and more cost-effective deployment of telecommunications services. The ETSI MANO framework plays a critical role in orchestrating these networks, yet it faces challenges such as the hidden state problem, arising from the NFVO's lack of holistic visibility into the internal state of NFVI-PoPs, which can lead to the choice of sub-optimal allocation schemes. This work introduces a novel approach to address the hidden state problem by integrating the Digital Twin (DT) paradigm into the MANO architecture. The proposed DT is a model-based solution employing neural networks to predict orchestration costs and estimate prediction errors, enabling the NFVO to make informed orchestration decisions through what-if analyses while preserving scalability and administrative independence. Performance evaluation demonstrates the DT's ability to mimic the behavior of an NFVI-PoP with high precision, i.e., in 84% of the cases, it returns a prediction that is 5% close to the actual value. Furthermore, the DT-aided NFVO achieves orchestration performance equivalent to approaches that assume full knowledge of the actual allocation costs, while overcoming in the 43% of cases traditional benchmark policies.

Modern enterprises depend on uninterrupted WideArea Network (WAN) connectivity between geographically dispersed sites and cloud services. Software-Defined Wide Area Networks (SD-WANs) have emerged as a compelling solution, offering greater flexibility and cost-efficiency compared to conventional WAN technologies. Ensuring robust availability-defined as the network's readiness to provide connectivity-is crucial for a successful SD-WAN implementation. This work presents a comprehensive, model-based evaluation framework designed to quantify service-oriented availability between pairs of SDWAN endpoints. Our hierarchical modeling approach effectively addresses the complexity and accurately reflects the dependencies among various SD-WAN components. Furthermore, we investigate the impact of a centralized control plane and its associated non-idealities on availability, providing insights into communication delays, sampling frequency of network characteristics, and algorithm execution times through simulation. In-depth experiments demonstrate the evaluation capabilities of our framework, including sensitivity analysis to identify critical components, the effects of various non-idealities, and how different control plane policies can mitigate them. These results offer valuable guidance for optimizing SD-WAN design before real-world implementation.

Software-Defined Wide Area Networks (SD-WANs) provide flexibility and cost-efficiency but still face significant challenges in guaranteeing high availability and consistent performance, particularly when utilizing diverse internet access networks, especially if best-effort like broadband connectivity. Furthermore, reliance on centralized controllers can introduce bottlenecks and single points of failure. This paper outlines a PhD research roadmap aimed at developing a resilient, performant, and extensible SD-WAN framework. The proposed research starts with robust modeling of service-oriented availability considering control plane impacts and quality of service requirements. This will guide us through an availability-aware SD-WAN modular architecture with the function of simplifying the development and test of new features such as: strategies to enhance edge autonomy by reducing real-time controller dependency; integration of predictive awareness tailored to specific internet access networks as LEO satellite based; and experimenting advanced monitoring for intelligent and reactive edgebased traffic engineering. We present the research context, state-of-the-art limitations, core research questions, the methodology combining modeling, simulation, and prototype development, and the expected contributions towards next-generation SD-WAN architectures suitable for research and demanding enterprise applications.

As artificial intelligence and machine learning (AI/ML) models become integral to network operations, their lack of transparency poses a significant barrier to operator trust. Existing explainable artificial intelligence (XAI) techniques often fail to bridge this gap for non-specialists, producing technical outputs that are difficult to translate into actionable insights. This paper presents a framework specifically designed to address this shortcoming. It leverages a moderately sized large language model (LLM) and extends beyond the standard use of SHapley Additive exPlanations (SHAP) feature influence values. The framework employs a structured prompt enriched with mutual feature interaction data to generate human-understandable natural language explanations. To validate our framework, we performed an empirical evaluation on an optical quality of transmission (QoT) estimation use case with human evaluators. We collected independent performance evaluations from specialists, which showed a high inter-evaluator agreement. Compared to a state-of-the-art baseline that uses only SHAP feature influence values in a straightforward prompt, our approach improves the explanation usefulness and scope by$\mathbf{1 2. 2 \%}$and$\mathbf{6. 2 \%}$, while achieving 97.5% correctness.

Software-Defined Wide Area Networks (SD-WANs) have emerged as a transformative solution for modern enterprise networking, enabling dynamic traffic management, cost-efficient connectivity, and improved network performance. However, ensuring real-time visibility into network conditions remains a key challenge, as SD-WAN overlay tunnels operate over diverse and often unpredictable underlay networks. Traditional network monitoring techniques, such as active and passive monitoring, face limitations in balancing accuracy, responsiveness, and overhead. To address this challenge, we propose an In-Band Network Telemetry (INT) framework for SD-WANs, leveraging extended Berkeley Packet Filter (eBPF) technology for efficient and flexible packet processing. Our approach enables real-time telemetry data collection at the Customer Premises Equipment (CPE) level, allowing for precise performance monitoring while minimizing additional network overhead. The framework integrates seamlessly with various VPN-based SD-WAN tunnels, including Generic Routing Encapsulation (GRE), IP Security (IPSec), and IPSec over GRE, ensuring adaptability across different deployment scenarios. By embedding telemetry metadata directly into overlay packets, the proposed solution provides continuous monitoring of critical Quality of Service (QoS) metrics, such as One-Way Delay (OWD), Two-Way Delay (TWD), and packet loss rate. Through extensive experimentation, we demonstrate the effectiveness of our INT-enabled SD-WAN framework in accurately detecting network anomalies and ensuring Service-Level Agreement (SLA) compliance. The results validate our approach as a scalable and lightweight monitoring solution for enhancing network observability in SD-WAN deployments.

The rapid increase in connected devices has significantly intensified the computational and communication demands on modern telecommunication networks. To address these challenges, integrating advanced Machine Learning (ML) techniques like Federated Learning (FL) with emerging paradigms such as Multi-access Edge Computing (MEC) and Software-Defined Wide Area Networks (SD-WANs) is crucial. This paper introduces online resource management strategies specifically designed for FL model aggregation, utilizing intermediate aggregation at edge nodes. Our analysis highlights the benefits of incorporating edge aggregators to reduce network link congestion and maximize the potential of edge computing nodes. However, the risk of network congestion persists. To mitigate this, we propose a novel aggregation approach that deploys an aggregator overlay network. We present an Integer Linear Programming (ILP) model and a heuristic algorithm to optimize the routing within this overlay network. Our solution demonstrates improved adaptability to network resource utilization, significantly reducing FL training round failure rates by up to 15% while also alleviating cloud link congestion.

Wide Area Networks (WANs) are essential for enabling communication among different branches within enterprises, ultimately improving operational efficiency and fostering economic advancement. However, the traditional WAN solutions often struggles to meet the diverse needs of geographically dispersed organizations. The advent of Software-Defined Wide Area Network (SD-WAN) solutions represents a paradigm shift, harnessing Software-Defined Networking (SDN) principles to transcend the limitations inherent in traditional WAN architectures. However, the necessity for advanced monitoring solutions persists as fundamental in guaranteeing optimal performance and unwavering reliability. This paper proposes the integration of In-band Network Telemetry in SD-WANs through Extended Berkeley Packet Filter (eBPF) technology to assess network performance parameters, such as the delay, in real-time. This work addresses the increasing demand for performance, security and reliability in such network architectures by leveraging INT capabilities for monitoring and manipulating data packets. Through the exploitation of eBPF’s features and network programmability, we introduce a novel approach aimed at measuring the end-to-end delay across SD-WAN deployments.

In the context of the ever-evolving 5G landscape, where network management and control are paramount, a new Radio Access Network (RAN) as emerged. This innovative RAN offers a revolutionary approach by enabling the flexible distribution of baseband functions across various nodes, all tailored to meet the ever-shifting demands of both system requirements and user traffic patterns. As users move within the network, the need to anticipate and strategically position these baseband functions becomes crucial for seamless network operation. Traditionally, this challenge has been tackled through a two-step process: first, forecasting traffic patterns, and then optimizing resource allocation accordingly. However, this approach falls short in guaranteeing an efficient placement when actual traffic demands surge onto the network. It often leads to resource overbooking, constraint violations, and excessive power consumption, putting strain on the network’s capabilities. In this paper, we introduce a novel framework based on a black-box optimization approach. This tool empowers prediction algorithms not just with historical traffic data but also with insights from optimization outcomes. The goal is to minimize a loss function related to power consumption and constraint violation: this ensures a predicted placement that is feasible and whose power is close to optimal. This approach ensures that the predicted placement is both feasible and power-efficient, bridging the gap between theoretical prediction and practical implementation. Remarkably, our proposed method, while potentially sacrificing some degree of traffic prediction accuracy, outperforms the conventional two-step approach by delivering a more efficient baseband function placement.

Deep Reinforcement Learning (DRL) is being investigated as a competitive alternative to traditional techniques for solving network optimization problems. A promising research direction lies in enhancing traditional optimization algorithms by offloading low-level decisions to a DRL agent. In this study, we consider how to effectively employ DRL to improve the performance of Local Search algorithms, i.e., algorithms that, starting from a candidate solution, explore the solution space by iteratively applying local changes (i.e., moves), yielding the best solution found in the process. We propose a Local Search algorithm based on lightweight Deep Reinforcement Learning (DeepLS) that, given a neighborhood, queries a DRL agent for choosing a move, with the goal of achieving the best objective value in the long term. Our DRL agent, based on permutation-equivariant neural networks, is composed by less than a hundred parameters, requiring only up to ten minutes of training and can evaluate problem instances of arbitrary size, generalizing to networks and traffic distributions unseen during training. We evaluate DeepLS on two illustrative NP-Hard network routing problems, namely OSPF Weight Setting and Routing and Wavelength Assignment, training on a single small network only and evaluating on instances 2x-10x larger than training. Experimental results show that DeepLS outperforms existing DRL-based approaches from literature and attains competitive results with state-of-the-art metaheuristics, with computing times up to 8x smaller than the strongest algorithmic baselines.

The continuous growth of advanced services integrating processing, storage, real-time data exchange, and transmission capacities highlights the importance of Quality of Service (QoS) and Quality of Experience (QoE). In particular, it is very relevant for enterprises that rely on Wide Area Networks (WANs) for reliable and efficient communication between their headquarters and branches. With the evolution of new applications with stringent requirements, ensuring high-performing WANs is a critical priority for businesses looking to remain competitive and provide a seamless customer experience. This paper explores using Rein-forcement Learning (RL) algorithms on Software-Defined Wide Area Network (SD-WAN) to improve QoS and reduce costs. SD-WAN allows for the dynamic reconfiguration of network devices in real-time, better meeting network measurements and service requirements. By leveraging self-learning techniques such as RL, which exploits feedback mechanisms, we can improve network availability by automatically routing traffic over existing network technologies. We also compare two SD-WAN topology scenarios, including direct WAN connections between Customer Premises Equipment (CPEs) within enterprise premises and CPEs used as peering points for traffic routing. Our approach shows promising results regarding network performance and cost-effectiveness, which can benefit businesses looking to improve their network infrastructure.

Quality of Service (QoS) and Quality of Experience (QoE) are the most relevant requirements for new advanced services comprising integrated cooperation between processing, storage, sensing, and transmission capabilities. As reliance on technology grows, businesses increasingly turn to Wide Area Networks (WANs) connectivity solutions to ensure reliable and efficient communication between their locations and cloud data centers. With new applications with stringent requirements, QoS and QoE have become critical priorities for companies looking to remain competitive and provide a seamless customer experience. As a result, many organizations require reliable and high-performing WANs to effectively transmit critical data between their branches and cloud data centers. We examine Multi-Protocol Label Switching (MPLS technology), which is the historical choice for WANs but has significant disadvantages in cost and performance. Then, we analyze the novel Software-Defined Wide Area Network (SD-WAN) technology that allows for real-time flexible, dynamic reconfiguration of network devices to meet network measurements and service requirements. Furthermore, we introduce a first work of my Ph. D. that examines the use of a Reinforcement Learning algorithm in a new SD-WAN topology scenario in which we consider both direct WAN connections between Customer Premises Equipment (CPEs) located within enterprise premises and CPEs used as peering points for traffic routing. This paper presents our initial analysis of these technologies and the main ideas guiding me in my Ph. D. program.

Network slicing is an important characteristic of 5G/6G networks that increases flexibility and enables different applications over a single infrastructure. The physical resources are partitioned to create virtualized networks, each dedicated to services with specific requirements. Several entities participate in network slicing, including Mobile Network Operators (MNOs), Mobile Virtual Network Operators (MVNOs), and users. An MNO owns the physical network infrastructure and the resources. MVNOs lease resources from the MNO and operate as service providers towards their subscribers. The goal of this work is to optimize the end-to-end network slicing process to provide services to users with a fair sharing of resources. We model this problem as a hierarchical combinatorial auction with a modified Vickrey-Clarke-Groves pricing mechanism. In the upper-level auction, an MNO is the seller supplying Network Slice to several MVNOs, who act as the bidders. In the lower-level auction, each MVNO holds an auction as a seller delivering services to their subscribed end-users, who play the role of bidders. We formulate and solve the Winner Determination Problem using mathematical programming and heuristic algorithms. The simulations show that the model can achieve fair sharing of resources, and it enables improving the MNO and MVNO revenue.

The increased demand for high-quality Internet connectivity resulting from the growing number of connected devices and advanced services has put significant strain on telecommunication networks. In response, cutting-edge technologies such as Network Function Virtualization (NFV) and Software Defined Networking (SDN) have been introduced to transform network infrastructure. These innovative solutions offer dynamic, efficient, and easily manageable networks that surpass traditional approaches. To fully realize the benefits of NFV and maintain the performance level of specialized equipment, it is critical to assess the behavior of Virtual Network Functions (VNFs) and the impact of virtualization overhead. This paper delves into understanding how various factors such as resource allocation, consumption, and traffic load impact the performance of VNFs. We aim to provide a detailed analysis of these factors and develop analytical functions to accurately describe their impact. By testing VNFs on different testbeds, we identify the key parameters and trends, and develop models to generalize VNF behavior. Our results highlight the negative impact of resource saturation on performance and identify the CPU as the main bottleneck. We also propose a VNF profiling procedure as a solution to model the observed trends and test more complex VNFs deployment scenarios to evaluate the impact of interconnection, co-location, and NFV infrastructure on performance.

Today, more and more enterprises are embarking on a digital transformation where most of their applications are hosted in the Cloud. As a result, a reliable Wide Area Network (WAN) has become a primary need to interconnect their distributed branch offices and data centers that accommodate those applications. Software-Defined Wide Area Network (SD-WAN) represents the most promising technology solution for next-generation enterprise networks, being able to increase network agility and reduce costs. In this paper, we present an experimental SD-WAN solution capable of running and optimizing delay-sensitive high-priority services, such as real-time video streaming, while minimizing downtime caused by network failures. This solution comprises a monitoring and a traffic engineering system for SD-WAN. The first consists of a Transport-layer Passive Monitoring (TPM) system based on extended Berkeley Packet Filter (eBPF) technology with the goal of monitoring TCP flows; the second consists of an application, running inside the SD-WAN controller, with the goal of orchestrating the network traffic in consideration of the monitoring measurements by ensuring rapid recovery and resilience in case of unexpected congestion events. We validate our solution over two SD-WAN testbeds: the first is hosted in our laboratory at Politecnico di Milano, while the second is deployed in a municipal network of an Italian city. Results show that our SD-WAN solution can increase the overall service availability while meeting the stringent QoS requirements of delay-sensitive services.s

The 5G Radio Access Network (RAN) virtualization aims to improve network quality and lower the operator’s costs. One of its main features is the functional split, i.e., dividing the instantiation of RAN baseband functions into different units over metro-network nodes. However, its optimal placement is non-trivial: it depends on the application requirements and on the expected traffic volume, whose daily variation highly impacts the total power consumption. Current optimization solutions fail to provide a placement solution capable of handling traffic fluctuations. In fact, the standard machine learning algorithms used in the literature for planning the network resources in advance result in an allocation that is inadequate to carry the actual traffic at all the time-slots. Hence, we must reserve an artificial buffer capacity in the nodes to ensure feasibility. Instead, our proposed method exploits a fine-grained two-step multi-task algorithm that predicts the mean and quantile traffic, making the artificial capacity no longer necessary. The subsequent placement uses mixed-integer linear programming and a heuristic. The former considers the expected traffic in the objective function (to estimate costs) and the quantile in the constraints (to enforce capacity limits). The heuristic combines the mean and quantile results to minimize the power and comply with the requirements. While using sufficiently large artificial buffers guarantees robustness with a mild power increase compared to the oracle, the fine-grained multi-task model improves the results, reducing the power consumption compared to the mean and meets all constraints. The heuristic enables significant computational time reduction.

The demand for reliable and efficient Wide Area Networks (WANs) from business customers is continuously increasing. Companies and enterprises use WANs to exchange critical data between headquarters, far-off business branches and cloud data centers. Many WANs solutions have been proposed over the years, such as: leased lines, Frame Relay, Multi-Protocol Label Switching (MPLS), Virtual Private Networks (VPN). Each solution positions differently in the trade-off between reliability, Quality of Service (QoS) and cost. Today, the emerging technology for WAN is Software-Defined Wide Area Networking (SD-WAN) that introduces the Software-Defined Networking (SDN) paradigm into the enterprise-network market. SD-WAN can support differentiated services over public WAN by dynamically reconfiguring in real-time network devices at the edge of the network according to network measurements and service requirements. On the one hand, SD-WAN reduces the high costs of guaranteed QoS WAN solutions (as MPLS), without giving away reliability in practical scenarios. On the other, it brings numerous technical challenges, such as the implementation of Traffic Engineering (TE) methods. TE is critically important for enterprises not only to efficiently orchestrate network traffic among the edge devices, but also to keep their services always available. In this work, we develop different kind of TE algorithms with the aim of improving the performance of an SD-WAN based network in terms of service availability. We first evaluate the performance of baseline TE algorithms. Then, we implement different deep Reinforcement Learning (deep-RL) algorithms to overcome the limitations of the baseline approaches. Specifically, we implement three kinds of deep-RL algorithms, which are: policy gradient, TD- λ and deep Q-learning. Results show that a deep-RL algorithm with a well-designed reward function is capable of increasing the overall network availability and guaranteeing network protection and restoration in SD-WAN.

Nowadays networks are the basis of our communication providing a great number of services. As a consequence, the number of networks is increasing as well as the traffic routed over them. There is an increasing demand for new services that require stringent constraints on capacity, latency and jitter to provide an appropriate Quality of Service (QoS) to end users. In order to cope with those requirements, network infrastructure needs to evolve from a static and closed architecture towards a more scalable, dynamic and agile one. SDN and NFV allows to provide different services, each one with its own QoS constraints, independent and secure, thanks to the network slicing concept, the main subject of this work. Network slicing allows to segment the underlying physical network into different logical networks to provide data transport customized to specific services. In this paper, we propose and early stage work with the aim to provide two mathematical models able to dynamically provision network slices on the physical network, complying with their QoS and reducing the power consumption for their instantiation and routing of traffic.

Recently, machine learning (ML) has attracted the attention of both researchers and practitioners to address several issues in the optical networking field. This trend has been mainly driven by the huge amount of available data (i.e., signal quality indicators, network alarms, etc.) and to the large number of optimization parameters which feature current optical networks (such as, modulation format, lightpath routes, transport wavelength, etc.). In this paper, we leverage the techniques from the ML discipline to efficiently accomplish the routing and wavelength assignment (RWA) for an input traffic matrix in an optical WDM network. Numerical results show that near-optimal RWA can be obtained with our approach, while reducing computational time up to 93% in comparison to a traditional optimization approach based on integer linear programming. Moreover, to further demonstrate the effectiveness of our approach, we deployed the ML classifier into an ONOS-based software defined optical network laboratory testbed, where we evaluate the performance of the overall RWA process in terms of computational time.