Software-Defined Networking (SDN) enables flexible and programmable control over network behavior through the deployment of multiple control applications. However, when these applications operate simultaneously, each pursuing different and potentially conflicting objectives, unexpected interactions may arise, leading to policy violations, performance degradation, or inefficient resource usage. This paper presents a Digital Twin (DT)-based framework for the early detection of such application-level conflicts. The proposed framework is lightweight, modular, and designed to be seamlessly integrated into real SDN controllers. It includes multiple DT models capturing different network aspects, including end-to-end delay, link congestion, reliability, and carbon emissions. A case study in a smart factory scenario demonstrates the framework’s ability to identify conflicts arising from coexisting applications with heterogeneous goals. The solution is validated through both simulation and proof-of-concept implementation tested in an emulated environment using Mininet. The performance evaluation shows that three out of four DT models achieve a precision above 90%, while the minimum recall across all models exceeds 84%. Moreover, the proof of concept confirms that what-if analyses can be executed in a few milliseconds, enabling timely and proactive conflict detection. These results demonstrate that the framework can accurately detect conflicts and deliver feedback fast enough to support timely network adaptation.

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.

The rise of new applications, such as interactive remote presence, online gaming, and video-assisted remote control of industrial machinery, necessitates enhanced requirements in terms of throughput and delay stability. Many efforts have been made to address these needs, with Deterministic Networking (DetNet) being one such initiative. DetNet aims to guarantee delivery with low latency and minimal jitter, ensuring high reliability and performance for time-sensitive applications. However, DetNet applicability in real-world scenarios is limited due to the need of a lower-layer protocol supporting resource reservation procedures (e.g., MPLS), and the lack of publicly available implementations. In this work, we present SRv6 Live-Live, an easy-to-deploy and highly scalable implementation of DetNet functions using the Segment Routing over IPv6 (SRv6) model. The SRv6 Live-Live behavior replicates packets of a selected flow across multiple paths at the ingress of the SRv6 domain and drops redundant replicas at the egress. After discussing insights about the paths’ selection strategy, we provide a SRv6 Live-Live implementation for programmable data planes using P4. We also propose the use of SRv6 Live-Live for best path selection at line rate, in SD-WAN scenarios. The main results obtained in the extensive performance evaluation are that SRv6 Live-Live preserves the throughput in case of congestion and reduces the tail end-to-end delay with a marginal impact on best-effort flows.

Next-generation applications, marked by their critical nature, need to cope with stringent Quality of Service (QoS) requirements, such as low response time and high throughput. Moreover, the increasing number of devices connected to the Internet and the need to provide a consistent network infrastructure to serve the applications requested by users, open the tradeoff of jointly considering the QoS improvement for such applications and the reduction in the energy consumption of the infrastructure. To address this challenge, this paper proposes ELTO (Energy-Load Trade-Off), a system designed for the joint optimization of energy efficiency and traffic load balancing during the transition from IP networks to Software-Defined Networks (SDN). Leveraging SDN and Network Function Virtualization (NFV) paradigms, ELTO introduces an Integer Linear Programming multi-objective formulation, and a Genetic Algorithm heuristic to tackle the optimization problem in large-scale scenarios. ELTO encompasses a holistic approach to network configuration, including network equipment status and routing, to strike a balance between network traffic load balancing and energy efficiency. Results over realistic topologies show the effectiveness of the proposed solution, outperforming other state-of-the-art approaches, being able to switch off nearly half of the links in the network while also reducing the Maximum Link Utilization.

The network programming model of the Segment Routing (SRv6) architecture offers the possibility to define new functions aiming at improving the network performance. In this paper we introduce SRv6 Live-Live, a new behavior for the SRv6 data plane. SRv6 Live-Live is based on two primitives: i) traffic duplication, performed at the ingress node, and ii) the traffic de-duplication, executed at the egress node. The proposed behavior is suitable for the service provisioning of traffic flows having stringent requirements in terms of reliability, low delay and high throughput. Our preliminary performance evaluation, conducted in an emulated environment and realized by using a prototype implementation based on P4, shows that SRv6 Live-Live enhances the performance of the selected traffic flows in challenging network scenarios, characterized by high level of packet corruption/loss and large values of bandwidth-delay products.

In band Network Telemetry (INT) is a technique aiming at collecting telemetry information by inserting it inside the data packets, instead of relying on classical centralized monitoring elements that periodically query the network devices. The main drawback of INT is represented by the introduced per-packet overhead, that could negatively affect some traffic flows, especially those having stringent QoS requirements. To deal with the increase in the packet length caused by INT, in this paper we introduce the Sampling and Recovering paradigm to overcome the classical Collect Everything approach where all the INT data must be gathered. The proposed approach hinges on signal processing strategies to sample and recover sparse flow signals. The key idea is to reduce the number of INT data to collect and exploit signal reconstruction algorithms to obtain the unseen samples. The preliminary performance evaluation shows that the 18% of INT data are enough to get an accurate reconstruction of the overall network situation, while allowing for 90% of overhead reduction with respect to the Collect Everything case.

Nowadays, inter domain routing optimization is performed based on the so called “Tweak and Pray” approach, which consists in performing changes in the configuration of the BGP protocol without knowing in advance the consequences of such a modification. This is due to the lack of cooperation among Network Operators in the configuration of the BGP to optimize the inter domain routing. Inefficiency in the resource usage, network anomalies and outages are common consequences of wrong configuration changes performed by Network Operators in an attempt to improve the performance of their infrastructures. In this paper we propose a novel framework based on the Digital Twin technology to enable the execution of “what-if” analysis in the context of Traffic Engineering performed by tuning BGP parameters. Such a paradigm shift will allow Network Operator to be aware of the effects of BGP configuration changes before their actual execution. A proof of concept related to the balancing of inbound traffic in an Autonomous System network, based on the use of the AS Path Prepending technique, is realized to validate the feasibility of the proposed approach.

Network Black Holes (BHs) are logical failures that create a service disruption for a subset of traffic flows, generally due to device misconfiguration. Detection of a BH is a hard task due to its specific nature: the infrastructure is up and the disconnection affects a limited number of flows. An example of BH is the one caused by the failure of the Path MTU Discovery procedure in IPv6. The Segment Routing (SR) Architecture is an overlay infrastructure that provides source routing support by exploiting the connectivity service offered by the underlay IPv6 (SRv6). Thus, SR inherits the problems related to BHs affecting IPv6. In SR this problem is even more stressed due to the encapsulation mechanism that is required to enforce the segment lists on packets. Even worse, existing active probing based tools to detect network BHs for IPv6 are not suitable in SR. In this paper we investigate the problem of detecting SR Black Holes in SR domains. As first, we provide an experimental demonstration of the creation of an SR Black Holes. Then we show that existing tools based on active probing are not suitable to detect SR BHs. Then, a passive framework named Segment Routing Black Holes Detection (SR-BHD) is introduced. SR-BHD makes use of specific traffic counters available in SR capable nodes to verify the validity of the flow conservation principle on each network element. Experimental evaluation carried out through simulation and emulation shows the effectiveness of SR-BHD in detecting the presence of SR BHs.

In this paper we introduce QLR, a per-router control agent that aims at reducing the occupancy of the local buffers by performing re-routing operations. The Segment Routing architecture is exploited to manage the uncoordinated selection of re-routing performed by different nodes, thus avoiding the creation of routing loops, while the Extensible In-band Processing is used to allow the network nodes to have a detailed and updated view of the wide network status. Data and control plane programmability are considered to define a prototype implementation of QLR that allows for the execution of a preliminary performance evaluation and proof-of-concept. From the conducted experiments has emerged that QLR can effectively reduce the maximum queue occupancy and end-to-end delay up to 43% and 63%, respectively.

The Network Programming model of SRv6 allows the creation of network programs that can be enforced over traffic flows entering a Segment Routing (SR) domain. A network program is a list of instructions that must be applied on a packet traversing the SR domain. Instructions, also known as behaviors, currently available in SRv6 are divided into two main categories: i) topological (e.g., send the packet over the shortest path), and ii) service based (e.g., duplicate the packet). In this paper we introduce a new behavior for the SRv6 Network Programming model, named maximize Throughput (max_T). This function allows to steer an incoming traffic flow toward the egress node over the path that currently guarantees the highest throughput for the flow. The proposed max_T behavior has been implemented over programmable switches, and its effectiveness in improving the performance experienced by flows asking for its application is evaluated through experiments performed over an emulated environment. The preliminary result shows that a 23% reduction of the transfer time for a file over the SR domain is achieved when the max_T behavior is used.

In this work, we provide a Machine Learning framework for augmenting the Differentiated Services (DiffServ) protocol with fine-grained dynamic traffic classification. The framework is called L-DiffServ. It is composed of two classification algorithms able to detect the QoS classes of incoming packets only looking at three packet header fields; the first algorithm, referred to as Inter-L-DiffServ, is a semi-supervised classification procedure able to replicate DiffServ classification; the second one, referred to as Intra-L-DiffServ, is an unsupervised algorithm for intra-class classification, useful for classes taking large portions of the overall traffic. We apply the latter to the low priority best-effort class. The performance evaluation shows that our solution is able to dynamically classify packets and to detect new QoS sub-classes hence adapting to traffic aggregate characteristics. We also show that network resource management can be improved exploiting the new generated QoS sub-classes: two active queue management algorithms based on WRED and CHOKe show a reduction of the number of sessions affected by packet losses up to 40% with respect to the legacy DiffServ procedure.

Service Function Chaining (SFC) is an enabling technology to provide end-to-end service differentiation according to specific user requirements. Although emerging technologies such as Software-Defined Networking (SDN) and Network Function Virtualization (NFV) are perfect enablers for SFC, hardware limitation of Ternary-Content Addressable Memories (TCAMs) can be an obstacle when handling a large variability of SFC requests, derived from the increasing number of users, and the heterogeneity of applications and Quality of Service (QoS) requirements. This article introduces and investigates the problem of TCAM size limitation on the classification procedure of SFC requests in SDN-based SFC environments. To overcome this limitation, the classification of incoming SFC requests is proposed to be offloaded to transient nodes when the occupation of the ingress node flow table is close to its maximum. An Integer Linear Programming (ILP) formulation is provided to formalize the Chain Request Classification Offloading (CRCO) problem, that consists in maximizing the number of SFC requests that can be served. Furthermore, a heuristic algorithm is presented to solve the CRCO problem in feasible time. The performance evaluation carried out over two real topologies, shows that the proposed offloading strategy can greatly increase the number of accepted requests without significantly affecting the network QoS.

With the increasing tendency on data rates in forthcoming communication networks, availability is a crucial aspect to guarantee Quality of Service (QoS) requirements. The possibility of predicting the lifetime of networking hardware can be a key to improve the overall network QoS. This paper proposes a generic Machine Learning (ML) based framework that learns how to mimic the mathematical model behind the lifetime of network line cards. Results show that a good precision (85%) and recall (close to 100%) on the estimation can be achieved regardless the type of line cards the network is composed of.

The advent of the Software Defined Networking (SDN) paradigm represents a great opportunity for the definition of new network management solutions. In this work, we focus on the definition and implementation of a novel technique to solve the Traffic Matrix Assessment (TMA) problem from the perspective of an Internet Service Provider. Since the migration from legacy IP networks to fully-deployed SDN ones needs to be incremental due to budget and technical constraints, this paper proposes a mixed measurement and estimation scalable solution for hybrid IP/SDN networks to accurately solve the TMA problem by exploiting the availability of flow rule counters in SDN switches. The performance evaluation shows that our error-tolerant solution is able to assess the TM with a negligible estimation error by only measuring a small percentage of traffic flows, overcoming other state-of-the-art algorithms proposed in the literature. Moreover, the performance analysis of the proposed implementation using the OpenDaylight controller over an emulated network environment, shows that a trade-off between the quality of the assessed TM and its impact on the network in terms of control messages can be found by properly tuning the number of measured flows.

Self-driving networks represent the next step of network management techniques in the close future. A fundamental point for such an evolution is the use of Machine Learning based solutions to extract information from data coming from network devices during their activity. In this work we focus on a new type of data, available thanks to the definition of the novel SRv6 paradigm, referred to as SRv6 Traffic Counters (SRTCs). SRTCs provide aggregated measurements related to forwarding operations performed by SRv6 routers. In this work a detailed description of different SRTCs types (SR.INT, PISD, PSID.TM and POL) is provided and their relationships is formalized. The theoretical framework deployed is used to identify, on the basis of network configuration parameters of both SRv6 and IGP protocols, the minimum set of independent SRTCs to characterize the Network Status: we show that about the 80% of counters can be neglected with no information loss. We also apply our framework to two use cases: i) Traffic Matrix (TM) Assessment and ii) Traffic Anomaly Detection. For the TM assessment, we show that in a partially deployed SRv6 scenario a specific type of SRTCs, i.e., PSID, is more reliable than other ones; on the contrary, in a fully deployed scenario POL and PSID.TM counters provide the full TM knowledge. For the Traffic Anomaly Detection case, we show that known solutions based on link load measurements can be improved when integrating SRTCs information.

In this work we define a framework for the assessment of the Traffic Matrix (TM) of an Internet Service Provider (ISP) network. The solution, referred to as mrT, i) is completely distributed among network nodes, ii) is based on different Map-Reduce building blocks, iii) and is able to extract the Ingress-Egress nodes traffic relationships starting from raw traces captured on node interfaces. Each network node is able to compute a row of the TM having as input the local trace and low size files sent by other network nodes. mrT can be used for the TM computation of any packet switched network where the source and destination identifiers are unique and global, and computational resources are available in proximity of network nodes. The performance evaluation, carried out with synthetic and real traffic traces, highlight that mrT is a suitable solution for the assessment of the TM in real ISP networks with high traffic volumes. Moreover, the use of the Map-Reduce paradigm allows a reduction of more than the 50% of the execution times, with respect to an SQL-based approach.

Traffic matrix (TM) assessment is a key issue for optimizing network management costs and quality of service. This paper presents a method to measure the intensity of ingress-egress traffic flows on an Internet service providers network that overcomes the limits of the classical measurement-based approaches. The proposed algorithm, called segment routing perturbation traffic (SERPENT), uses a routing perturbation approach enabled by the segment routing paradigm: The paths of a subset of flows are changed so that their intensities can be determined measuring the variation of the load of the network links. The TM is measured in successive steps, called snapshots, in which sets of flows are progressively re-routed and measured, under a maximum link utilization constraint. We state an integer linear programming (ILP) optimization problem to determine the flows to be rerouted in one snapshot. SERPENT is an heuristic offering an efficient solution to the stated ILP. Results show that SERPENT assesses the intensity of more than 80% of flows even when the network is highly stressed, while reducing the configuration cost with respect to classical approaches. Moreover, when used in conjunction with an estimation algorithm, SERPENT allows a reduction of the estimation error by more than 50% with fewer than 5 snapshots.

We propose a framework to manage the link lifetime in an IP backbone network by exploiting the sleep mode (SM). In particular, when an SM feature is available, two different effects coexist: during the SM state, the lifetime tends to be increased and however, when the link changes its power state (from SM to full power or vice-versa), the lifetime tends to be decreased. We, therefore, define an optimal formulation of the lifetime-aware network problem. Moreover, we propose a heuristic, called Acceleration Factor Algorithm, to practically manage the device lifetime. We solve the problem both optimally and with our heuristic, considering two representative case studies. Results show that our approach outperforms the previous energy-aware algorithms, which instead do not consider the lifetime decrease triggered by the power state change. Thus, we argue that a lifetime-aware network management should be pursued when deciding to set an SM state for each device in an Internet Service Provider network.

Segment routing (SR) is a new routing paradigm to provide traffic engineering (TE) capabilities in an IP network. The main feature of SR is that no signaling protocols are needed, since extensions of the interior gateway protocol routing protocols are used. Despite the benefit that SR brings, introducing a new technology into an operational network presents many difficulties. In particular, the network operators consider both capital expenditure and performance degradation as drawbacks for the deployment of the new technology; for this reason, an incremental approach is preferred. In this paper, we face the challenge of managing the transition between a pure IP network to a full SR one while optimizing the network performances. We focus our attention on a network scenario where: 1) only a subset of nodes are SR-capable and 2) the TE objective is the minimization of the maximum link utilization. For such a scenario, we propose an architectural solution, named SR domain (SRD), to guarantee the proper interworking between the IP routers and the SR nodes. We propose a mixed integer linear programming formulation to solve the SRD design problem, consisting in identifying the subset of SR nodes; moreover, a strategy to manage the routing inside the SRD is defined. The performance evaluation shows that the hybrid IP/SR network based on SRD offers TE opportunities comparable to the one of a full SR network. Finally, a heuristic method to identify nodes to be inserted in the set of nodes composing the SRD is discussed.

Traffic matrix estimation in communication network is a long standing problem for its intrinsic difficulty and potential benefit to a vast number of network optimization and management functions. We address the improvement of the traffic matrix estimation by means of selected traffic flow measurements, besides the easily obtained link load measurements. The key contribution of this paper is the definition and assessment of an effective criterion, based on the flow spread parameter, to identify the flows to be measured that reduce the estimation error most. It turns out that a small percentage of flows are enough to drive the estimation error an order of magnitude lower than the one obtained with the classical solution solely based on link load measurements. Our algorithm, referred to as flow spread-based algorithm (FSBA), is also able to distribute measurement tasks fairly among network nodes, taking into account the available forwarding tables space. We also show that FSBA outperforms the state-of-the-art similar approaches. A detailed discussion on how the observation of the desired flows can be performed is done as well, by addressing the SDN paradigm. This is in fact a rapidly growing concept that enables individual flow measurements, though for only a limited number of flows to be practical with current technology.

In this work we study the impact of energy-aware routing algorithms on IP backbone networks, focusing on the routing protocol transients due to network reconfiguration. We first propose the Green Partial Exportation (GPE) algorithm, which is fully compatible with the OSPF protocol and targets the reduction of the number of changed paths in the network; we also realize a green software router by integrating GPE in the Quagga routing suite. Then we define an experimental methodology to evaluate the effects of a green routing strategy on the network behavior, in terms of delay increase and packet loss. Finally, we evaluate our solution on an emulated testbed from a national telecom operator. Our results show a maximum increment of 320 ms for the RTT and a packet loss of 1.45% during the network transients. Moreover, GPE can be safely applied in the network with a time granularity of less than one minute.

This paper proposes a novel energy-aware adaptive routing solution to face the energy consumption problem in IP networks, exploiting the IP over IP tunneling technique. The problem of minimizing the number of used links by configuring a set of tunnels, referred to as energy minimized tunneling (EMT), is modeled as an integer linear programming problem. The general EMT problem is then slightly modified by adding a set of constraints to make it more applicable to real networks scenario. With respect to previously proposed solutions, the EMT formulation exploits IP tunneling to change traffic routing assuring three main features: 1) no modifications to both data and control planes of current IP routers, so representing a feasible short-term solution for actual IP networks; 2) no need of IP routing protocols convergence to modify the routing paths, so avoiding routing instabilities; and 3) performance comparable with flow-based solutions requiring the deployment of a connection-oriented protocol. We propose a practical solution, called Virtual Interface oN Off (VINO), that exploits the tunnel aggregation mechanism in order to reduce EMT operational complexity. VINO works in two steps and it is based on the preconfiguration of a limited set of (virtual) tunnel interfaces along with a set of static routes on IP routers; thus, these virtual interfaces are dynamically activated and deactivated according to traffic variations. The performance analysis carried out demonstrates that VINO is able to outperform IP green routing solutions proposed in the literature, in terms of energy efficiency.

Decreasing the soaring energy cost is imperative in large data centers. Meanwhile, limited computational resources need to be fairly allocated among different organizations. Latency is another major concern for resource management. Nevertheless, energy cost, resource allocation fairness, and latency are important but often contradicting metrics on scheduling data center workloads. Moreover, with the ever-increasing power density, data center operation must be judiciously optimized to prevent server overheating. In this paper, we explore the benefit of electricity price variations across time and locations. We study the problem of scheduling batch jobs to multiple geographically-distributed data centers. We propose a provably-efficient online scheduling algorithm - GreFar - which optimizes the energy cost and fairness among different organizations subject to queueing delay constraints, while satisfying the maximum server inlet temperature constraints. GreFar does not require any statistical information of workload arrivals or electricity prices. We prove that it can minimize the cost arbitrarily close to that of the optimal offline algorithm with future information. Moreover, we compare the performance of GreFar with ones of a similar algorithm, referred to as T-unaware, that is not able to consider the server inlet temperature in the scheduling process. We prove that GreFar is able to save up to 16 percent of energy-fairness cost with respect to T-unaware.

This paper deals with an energy saving routing solution, called Energy Saving IP Routing (ESIR), to be applied in an IP network. ESIR operation is integrated with Open Shorthest Path First (OSPF) protocol and allows the selection of the links to be switched off so that the negative effects of the IP topology reconfiguration procedures are avoided. The basic mechanisms which ESIR is based on are the concepts of SPT exportation and move. These mechanisms allow to share a Shortest Path Tree (SPT) between neighbor routers, so that the overall set of active network links can be reduced. Properties of moves are defined and the energy saving problem in an IP network is formulated as the problem of finding the Maximum Set of Compatible Moves (MSCM). The MSCM problem is investigated in two steps: firstly, a relaxed version of the problem, named basic MSCM problem, is considered in which QoS requirements are neglected; in the second step, the solution of the full problem, named QoS-aware MSCM problem, is faced. We prove that the basic MSCM problem can be formulated as the well-known Maximum Clique Problem in a graph; instead the QoS-aware MSCM introduces a condition equivalent to the Knapsack problem. ILP formulations to solve both the problems are given and heuristics to solve them in practical cases are proposed. The performance evaluation shows that in a real ISP network scenario ESIR is able to switch off up to 30% of network links by exploiting over-provisioning adopted by operators in the network resource planning phase and typical daily traffic trend.