Traditional SDN controllers are usually deployed as monolithic systems or tightly coupled service chains, limiting their adaptability in distributed or federated network domains. This paper presents eMSN, a microservice-based SDN framework that enables controller decomposition and decentralized multidomain orchestration, providing a foundation for scalable, and domain-aware SDN experimentation. The framework introduces lightweight, containerized microservices that interact via REST APIs and coordinate through a shared ETCD cluster. The main one, called FlowBlocker, collects topology and host information from the emitter, builds a policy-aware decision table, and shares domain-scoped state in ETCD. The latter stores only host/topology data and never flow rules; enforcement decisions remain within FlowBlocker, which installs rules locally via Ryu Core or coordinates with peer FlowBlockers across domains. The architecture supports centralized, partially decentralized, and fully decentralized deployment models. The proof-of-concept implementation uses Docker and Mininet for reproducibility. Functional evaluation demonstrates sub-millisecond Packet-In responsiveness, tens-of-milliseconds policy enforcement latency, and correct blocking of unauthorized traffic.

Software-Defined Networking traditionally relies on the separation of the control and data planes, centralizing network intelligence within a logically unified controller. However, centralizing control functionalities often introduces limitations that negatively impact the overall Quality of Service, particularly in distributed and heterogeneous network scenarios. In this paper, we explore an adaptive approach to orchestrate a microservice-based SDN controller dynamically at the Edge. Building upon our previously introduced frameworks for Microservice-based SDN Controller and for flexible service-model-aware orchestration, we investigate the benefits of adaptively deploying our microservice-based SDN controller’s functionalities at the Edge to enhance QoS. We leverage our orchestration framework to dynamically decide and execute the optimal placement of latency-critical microservices according to real-time monitoring data and evolving user demands. We evaluate the performance by comparing various deployment strategies, focusing on the tradeoff between control plane latency and placement of controller functionalities. Results demonstrate that our adaptive Edge deployment approach has the potential to reduces control plane latency, demonstrating the practical benefits of integrating SDN controller modularity with intelligent service orchestration in dynamic and heterogeneous network environments.

In SDN-enabled disaggregated optical networks, switches and wavelength converters are managed through software abstractions that model their capabilities, enabling greater flexibility and improved performance. While traditional approaches statically assign converters to specific nodes, such abstractions allow wavelength conversion to be considered as a distributed processing function, dynamically migrating across the network like a Cloud-based service. This enhances the ability of the control layer to flexibly allocate conversion resources to reduce blocking occurrences. Additionally, the orchestration layer dynamically selects wavelength allocation policies based on these abstractions, ensuring an optimal balance between performance and resource efficiency. This paper proposes a novel Cloud-native approach to wavelength conversion in disaggregated optical networks. We present a framework for orchestrating and controlling end-to-end services with a focus on lightpath provisioning, supported by a resource availability model that facilitates efficient wavelength selection. Our evaluation compares multiple wavelength assignment algorithms through both static and dynamic simulations. The results highlight key trade-offs among these algorithms in terms of blocking probability, resource utilization, and overall cost efficiency, providing valuable insights for improving the overall network performance.

Traditional Cloud-based multimedia transcoding approaches struggle with issues like latency and bandwidth limitations. In this work, we experiment with strategies that exploit Edge computing capabilities to get around such constraints. We focus on an on-demand multimedia transcoding service for real-time video streaming on the uplink, leveraging a state-of-the-art Edge Computing service orchestrator to handle its provisioning. We regard this as a use case for the orchestration platform, with the objective of enhancing the experience of the user as well as keeping important metrics in check, including resource utilization and load distribution over the Edge compute nodes. We perform evaluations on a physical testbed to assess the performance of the Edge-based system using real-world video transcoding workloads.

The Edge-Cloud Continuum represents a paradigm shift in distributed computing, seamlessly integrating resources from cloud data centers to edge devices. However, orchestrating services across this heterogeneous landscape poses significant challenges, as it requires finding a delicate balance between different (and competing) objectives, including service acceptance probability, offered Quality-of-Service, and network energy consumption. To address this challenge, we propose leveraging Multi-Objective Reinforcement Learning (MORL) to approximate the full Pareto Front of service orchestration policies. In contrast to conventional solutions based on single-objective RL, a MORL approach allows a network operator to inspect all possible “optimal” trade-offs, and then decide a posteriori on the orchestration policy that best satisfies the system’s operational requirements. Specifically, we first conduct an extensive measurement study to accurately model the energy consumption of heterogeneous edge devices and servers under various workloads, alongside the resource consumption of popular cloud services. Then, we develop a set-based MORL policy for service orchestration that can adapt to arbitrary network topologies without the need for retraining. Illustrative numerical results against selected heuristics show that our MORL policy outperforms baselines by 30% on average over a broad set of objective preferences, and generalizes to network topologies up to 5x larger than training.

We consider the problem of designing and training a neural network-based orchestrator for fog computing service deployment. Our goal is to train an orchestrator able to optimize diversified and competing QoS requirements, such as blocking probability and service delay, while potentially supporting thousands of fog nodes. To cope with said challenges, we implement our neural orchestrator as a Deep Set (DS) network operating on sets of fog nodes, and we leverage Deep Reinforcement Learning (DRL) with invalid action masking to find an optimal trade-off between competing objectives. Illustrative numerical results show that our Deep Set-based policy generalizes well to problem sizes (i.e., in terms of numbers of fog nodes) up to two orders of magnitude larger than the ones seen during the training phase, outperforming both greedy heuristics and traditional Multi-Layer Perceptron (MLP)-based DRL. In addition, inference times of our DS-based policy are up to an order of magnitude faster than an MLP, allowing for excellent scalability and near real-time online decision-making.

With the establishment of the Everything-as-a-Service (XaaS) paradigm for service provisioning, coupled with the increasingly-demanding requirements imposed by modern network services, the need for a XaaS-aware orchestration system able to cope with a heterogeneous infrastructure, such as the one of Fog Computing environments, is evident. In this work, we describe the working principles and implementation aspects that allow the orchestration of services offered according to the Function-as-a-Service (FaaS) model. The live demonstration will showcase the ability of the system to deploy this kind of services on a suitable test bed, with comments on the procedure and the performance.

Fog Computing can facilitate the adoption of the Everything-as-a-Service paradigm in infrastructure segments that are located closer to the end user, or to the data source, compared to typical Cloud solutions. This enables combining the advantages of flexible service deployment models with the need to cope with the strict requirements–especially in terms of latency–of emerging applications in softwarized networks. Along comes the need to consider aspects of service orchestration specific to the Fog environment and its intrinsically dynamic nature. In this paper we propose an architecture for flexible Fog Computing service orchestration, with a particular focus on the awareness of service deployment models. We discuss the design choices and describe the components and operations of the proposed orchestration system. We then present a complete working implementation of such architecture, including insights on its ability to handle critical orchestration functions such as service discovery and resource monitoring. We also report on the experimental validation of the system and the performance evaluation on real-world equipment, proving the feasibility and the effectiveness of the approach on a dynamic Fog infrastructure. We complement the work by presenting the results of a combinatorial analysis, validated by simulation, of the service model-aware resource selection process. As a result of our investigation, we show that Fog services can be effectively deployed in a matter of a few seconds, or even in less than one second when suitable Fog nodes are available, taking advantage of the awareness of the available service models.

Optical aggregation network is a key subsystem to support high performance end-to-end services in 5G mobile networks. This paper presents a network design optimization methodology to implement ultra reliable low latency communications (URLLC) in a metro area. Novel optimization algorithms based on Integer Linear Programming (ILP) are defined to solve dedicated (DPP) and shared (SPP) path protection problems, in the presence of single failure with the aim to minimize the number of active nodes, by adopting the functional splitting options defined by 3GPP. The results prove the ability of the proposed algorithm to find optimal solutions which also minimize the number of high capacity node instances and the usage of bandwidth resources, especially when SPP is applied.

Base station (BS) densification is increasingly adopted by mobile operators in order to support increasing traffic demand. However, a large number of BSs requires many backhaul connections, which may be very expensive. For this reason, provisioning high speed backhaul connections to BSs in a cost-effective way is challenging, and it is important to efficiently utilize an existing fixed network infrastructure if possible. This paper proposes two optimized infrastructure-aware planning strategies for small cells and fiber backhaul. The first strategy is referred to as joint design (JD) and is based on the joint cost minimization of small cells and fiber backhaul. The JD strategy is compared to a traditional design (TD) solution based on a two-step optimization approach. In the latter a cost-optimal small cells placement is identified first, then the corresponding minimum cost fiber backhaul deployment is determined accordingly. A comparison between these two approaches in dense 5G urban scenarios shows that the relative performance of JD and TD largely depends on the expected traffic demand and on the existing infrastructure. In dense urban areas with the average traffic levels expected for the year 2020 and beyond, JD returns up to 50% lower deployment cost in a greenfield scenario, and up to 70% lower deployment cost in a brownfield scenario. Only in areas with extremely high traffic demand (e.g., open-air festivals and stadiums) JD returns deployment costs very similar to TD.

Providing high capacity to the end users is one of the main challenges for the fifth generation (5G) of mobile networks. The users' habit to consume online contents indoor makes the outdoor-to-indoor capacity provisioning impractical, especially when the high-frequency bands proposed for 5G are employed. The centralized radio architecture (CRA) is an in-building solution, which relies on the centralization of baseband processing functions, fully or partly allowing for centralized cell management while providing signals directly inside the buildings. On the other hand, the massive deployment of CRAs in urban areas may yield to unacceptably high installation costs, due to the radio network equipment to be activated. To make CRAs appealing to mobile operators, we propose different deployment strategies to minimize the CRA deployment cost. We define the remote radio unit placement (RRUP) problem and formulate it as an integer linear program, obtaining optimal deployment solutions in small urban residential scenarios. We prove the RRUP problem to be NP-hard, requiring heuristic approaches to solve large problem instances. To this end, we propose an effective and scalable heuristic for minimizing the amount of radio equipment required to deploy CRAs in large urban areas.

In this paper a hierarchical architecture for content distribution in vehicular network is proposed. A cluster-based organization of vehicles is considered based on the Named Data Networking (NDN) model where a subset of reference vehicles named barycenters plays the NDN router role. Complementary functionalities of both short-range and cellular communications are used, the former for vehicle-to-vehicle (V2V), the latter for vehicle-to-infrastructure (V2I) communications, to off-load the fixed NDN infrastructure, and increase the chance of finding a content in the VANET, while limiting the number of broadcasted information. The proposed model is described and analytical evaluations are presented to show the potential of the approach.

The development of software defined networking (SDN) has instigated a growing number of experimental studies which demonstrate the flexibility in network control and management introduced by this technique. Optical networks add new challenges for network designers and operators to successfully dimension and deploy an SDN-based in the optical domain. At present, few performance evaluations and scalability studies that consider the high-bandwidth of the optical domain and the flow characterization from current Internet statistics have been developed. In this paper these parameters are taken as key inputs to study SDN scalability in the optical domain. As a relevant example an optical ring Metropolitan Area Network (MAN) is analyzed with circuit and packet traffic integrated at the wavelength level. The numerical results characterize the limitations in network dimensioning when considering an SDN controller implementation in the presence of different flow mixes. Employing flow aggregation and/or parallel distributed controllers is outlined as potential solution to achieve SDN network scalability.

This paper addresses contention resolution in optical switching fabrics exploiting Parametric Wavelength Converters (PWCs). Being a PWC able to perform simultaneous wavelength conversions, it can be employed to replace many single-channel wavelength converters, thus saving in the number of expensive optical components. Recently a switching fabric employing FixedWavelength Converters (FWCs), providing relevant cost savings compared to previous proposals, has been presented. The drawback is that it requires a high number of FWCs. For this reason, here new solutions based on PWCs instead of FWCs for such an architecture are proposed. Resulting switching fabrics are compared in terms of loss performance and cost. Simulation results outline that fabrics based on PWCs allow to save devices while obtaining the same performance as architectures using FWCs. They also prove how cost benefits are obtained in many switch configurations.

This paper presents a new hybrid network architecture including different transport technologies to support a wide range of services. In particular the proposed hybrid network provides three service levels, and a possible set of foreseeable services with different QoS needs is mapped into these levels to show the effectiveness of the hybrid networking in managing multi-service requests. In the switching nodes of the network, each service level is associated to a different switching technology, thus optimizing the switch implementation. In the nodes of the network, the service levels are distinguished based on optical encoding techniques, while an output collision stage manages the access to the output wavelengths. Performance studies show the effectiveness of the output collision mechanism.

In this paper, a buffer-less shared-per-wavelength optical switch is equipped with multi-fiber interfaces and operated in asynchronous context. An analytical model to evaluate loss performance is proposed using an approximate Markov-chain based approach and the model is validated by simulations. The model is demonstrated to be quite accurate in spite of the difficulty in capturing correlation effects especially for small switch sizes. The model is also applied to calculate the number of optical components needed to design the optical switch according to packet loss requirements. The impact of the adoption of multiple fiber interfaces is outlined in terms of the remarkable saving in the number of wavelength converters employed, while increasing at the same time the number of optical gates needed by the space switching subsystem. The numerical results produced are a valuable basis to optimize overall switch cost.

Dynamic, flexible and programmable network nodes will enable the key infrastructure for the Future Internet. Reengineering of the traditional router architectures is required in order to allow hardware- and vendor-independent routing functions and to design open and cost-effective service platforms. This paper presents a modular programmable router architecture to support multi-granular routing based on optical switching matrices, according to recent IETF standards. A possible implementation of the proposed concepts is demonstrated using a software-based optical router emulator, which integrates optical packet and circuit switching paradigms. The emulator is designed as a test platform to prove the feasibility of different subsystems and their interoperability.

This paper considers simple fixed-output wavelength converters to implement shared-per-node optical packet switching with wavelength domain contention resolution capability. Scheduling algorithm and logical performance analysis are presented in synchronous context. Packet loss probability is evaluated and validated through simulation, and results are applied for comparisons with previously proposed sharing schemes. Power consumption evaluation is also presented. The results obtained outline that the proposed architecture performs very close to the general shared-per-node scheme, while employing simpler components.

Programmable optical networks are a key solution to high performance dynamic network services in Future Internet scenario. Modular router architecture which meets this concept is here defined and represented in a flexible software context, using Click! tools. Evaluation of logical performance and testing of physical characteristics of information forwarding within this framework is shown as feasible. The approach allows also to easily test functions and interactions of control and data planes to support enhanced future Internet services. Sample implementations of programmable router subsystems are given and discussed.

This paper studies TCP performance with different offset times which arise when hybrid multi-granular technology is employed in OBS core nodes. Multi-granular switches are promising devices to match dynamic application needs while optimizing switch costs and feasibility. They foster optical burst sw

This paper describes an optical switching fabric which solves contention in a hybrid context, by exploiting both time and wavelength domains through wavelength converters and electronic queues. The relationship between the hybrid nature of the switch and the quality of service requirements of traffi

This paper presents a programmable multi-granular optical cross connect (MG-OXC) and network architecture deployable in multi-service and multi-provider networks. The concept of programmable MG-OXC is introduced to provide a way of utilizing multiple switching/transport granularities to efficiently

Most of the previous works on optical burst switching (OBS) assume in their analysis that signalling does not affect network performance. It is analysed here, under which conditions the effect of signalling is actually negligible, taking into account the effect of signalling in the evaluation of burst discard probability. First, analytical models for two different signalling approaches in an OBS network are presented: ‘out-of-band’ and ‘in-band’ techniques. The impact of these two signalling strategies in terms of the probability of burst discard are evaluated, identifying the component of bursts discarded as a consequence of control message losses or of excessive signalling delay. A new method is also discussed, based on the previous models, to assign the correct amount of resources to the control plane. To verify the accuracy of the analytical results, these are compared with results based on discrete-event simulationns: results are found to be in a highly satisfactory agreement with simulations.

In optical burst switched (OBS) networks, the queueing delay in ingress edge node is an important performance measure. In this paper, we propose a deterministic delay model and derive the upper bound of the burst queueing delay in an edge node. On this basis, the edge-to-edge delay guarantee framework for premium services is further outlined.

The wireless access scenario has a key role in today communication systems. On the other hand, high speed network paradigms like optical burst-switched networks are emerging as changeling technologies for next generation Internet. In this context, the Transmission Control Protocol (TCP) still remain the most deployed transport protocol. The assembly algorithm performed at the edge nodes to create optical bursts influences TCP congestion control in the presence of burst and single segment loss. Here this influence is analyzed in the heterogeneous scenario where wireless access networks feed an optical burst switched link. The results, obtained by ns-2 simulations, show that the burst assembly algorithms positively affect average throughput and TCP fairness with both size-based and time-based assembly algorithms.

This work considers contention resolution in optical packet switches and proposes a switch architecture empowered by available technology which exploits a combination of wavelength and time domains to solve packet contention. An hybrid switch architecture based on space switching matrix and re-circulating links, equipped with wavelength converters and electronic buffers is described. The forwarding procedure is implemented by suitable scheduling which minimizes the number of wavelength conversions needed. Switch performance is evaluated by simulation to show the effects of combination of wavelength conversion and electronic queuing. The hybrid switch characteristics are exploited to design the switch in relation to application needs. To this end, service differentiation schemes are applied to traffic forwarding and evaluated. The results obtained outline how switch design optimization take advantage of the relationships between the employed technology and traffic needs.

Synchronization of multiple TCP flows in optical burst switched networks is considered in this paper. TCP flow synchronization happens when concurrent flows accessing the optical network at the same edge node increase/decrease their congestion windows at the same time intervals, thus causing a significant instability of data transmission rate. Different burst assembly algorithms sensibly influence this phenomenon and are here studied to quantitative assess their effect. A simulation tool based on ns-2 is developed and applied to different case studies. A new burst assembly scheme is proposed to reduce flow synchronization and its effectiveness is evaluated.

In this paper we propose an edge-to-edge closed-loop scheme to provide Quality-of-Service (QoS) in Optical Burst Switching (OBS) networks. Performance parameters for each burst flow are exchanged by a feedback message called Service Control Message (SCM) which is processed only at the edge/ingress nodes. According to the information contained in the SCM the edge nodes could adapt the values of some flow parameters (typically burst assembly parameters and offset time) to apply differentiated quality of service to the various flows. Simulations results are presented to investigate the feasibility of this new proposal.

In this paper, we study an asynchronous optical packet switching node equipped with a number of limited range or full range wavelength converters shared per node. The packet traffic is realistically modeled by a superposition of a finite number of on-off sources as opposed to the traditional Poisson model which ignores the limited number of ports on a switch. We both study circular and non-circular limited range wavelength conversion schemes. In our simulations, we employ the far conversion policy where the optical packet is switched onto the farthest available wavelength in the tuning range, which is known to outperform the random conversion policy. We propose an approximate analytical method based on block tridiagonal Markov chains and fixed point iterations to solve for the blocking probabilities in share per node wavelength conversion systems. The method provides an accurate approximation for full range systems and acceptable results for limited range systems.

Next generation optical network technologies such as OPS and OBS have a non-negligible impact on the performance of the transport layer. This is related to how a packet stream traversing the optical network is affected by random behaviors such as latency variability, out-of-sequence delivery and loss. This paper tries to better understand how the use of contention resolution schemes at the optical layer affects the TCP performance, focusing in particular on the impact of unordered delivery and loss of packets.

This paper describes the scheduling algorithms to control the optical packet forwarding in a multi-stage switch based on the broadcast-and-select principle. The aim of control algorithms is to fully exploit switch resources while avoiding contention in the wavelength and space domains. A multi-stage switch based on the shared per node wavelength conversion scheme is considered in a synchronous context. An analytical model is provided to calculate the packet loss probability in relation to different reference traffic scenarios. Results show how the sharing of wavelength converters impacts on node performance depending on the scheduling algorithm applied, and provide a meaningful support for optical packet switch design.

This paper analyses an optical packet switch with partial wavelength conversion on output fibres, meaning that a reduced set of full range tuneable wavelength converters is used for contention resolution. The aim is to find a trade off between cost and performance. A simple analytical model is developed to evaluate packet loss probability and throughput. Numerical results, validated by simulation, are presented to show the influence on performance of different design parameters and the significant cost saving obtained while achieving the same packet loss rate of a fully equipped switch.

This papers reports the advances in optical core networks research coordinated in the framework of the e- photon/ONe and e-photon/ONe+ networks of excellence.

This paper presents an analytical model to calculate the send rate of TCP sources as a function of the main design parameters of optical burst-switched (OBS) networks. The novelty of the work is the possibility to deal with generic source speed that was previously evaluated only by simulation. The pretty good accuracy of the model is proved by comparisons with simulation results. The practical significance of the model is shown by an application example where burst loss is consequence of contention in OBS core nodes.

This paper addresses the problem of building optical packet switches that are able to effectively cope with variable length packet traffic and quality of service management, therefore able to support IP traffic. The paper aims at showing that the availability of dense wavelength division multiplexing is crucial. By suitably exploiting the wavelength dimension a multistage fiber delay line buffer can be implemented, with fine granularity and long delay with an architecture of limited complexity. This is necessary to fulfill the buffering requirements of variable length packets. Furthermore, the wavelength domain is proved to be more effective than the time domain to manage different levels of quality of service. Algorithms are presented that are peculiarly designed for this environment showing that they can effectively differentiate the packet loss probability between three priority classes.

The priority queueing mechanism is analysed to verify its effectiveness when applied for the support of expedited forwarding-based services in the differentiated services environment. An experimental measurement-based methodology is adopted to outline its properties and end-to-end performance when supported in real transmission devices. A test layout has been set up over a metropolitan area for the estimation of the one-way delay. The effect of the buffering architecture and of the background traffic have been evaluated, moreover a worst case analysis of the average one-way-delay that validates the experimental results is proposed.

This paper addresses the design of an optical packet switch able to effectively cope with variable length packet traffic, such as IP traffic. A switching architecture equipped with a multistage fiber delay line buffer is presented, that is able to realize fine time granularity and long delay. WDM is introduced to solve switch internal blocking and to enhance buffer exploitation. Packet loss performance, evaluated by simulation, is discussed in relation to the degrees of freedom available for switch design, to show the feasibility of a switch for the IP environment.

This paper investigates the overall traffic performance of a transparent optical network, suitable for high bit rate interconnection. The motivation of the introduction of such a transport network are discussed and the network architecture is presented. The performance is investigated in an end-to-end perspective. On one side the traffic resulting at the output of the interface with other networks, such as ATM for instance, is studied showing an interesting traffic shaping effect. Then this traffic is taken as the input to optical switches showing that the target performance in terms of the packet loss can be reached with present technology both in the access as well as in the core part of the network. An approximate evaluation of the end-to-end delay experienced by data crossing the optical network in a typical operating environment is given showing the effectiveness of the approach presented.

This paper reviews the work carried out in the ACTS KEOPS (Keys to Optical Packet Switching) project, describing the results obtained to date. The main objective of the project is the definition, development, and assessment of optical packet switching and routing networks, capable of providing transparency to the payload bit rate, using optical packets of fixed duration and low bit rate headers in order to enable easier processing at the network/node interfaces. The feasibility of the KEOPS concept is assessed by modeling, laboratory experiments, and testbed implementation of optical packet switching nodes and network/node interfacing blocks, including a fully equipped demonstrator. The demonstration relies on advanced optoelectronic components, developed within the project, which are described.

Input-buffered replicated networks are considered for broadband switching applications. They are characterized by many design parameters such as the replication factor, the traffic management policy, and input buffer location and length. To show the influence of these parameters on switching performance, an analytical model is defined based on a Markov chain representation of the input buffer. This model is suitable for application to input buffered architecture having different routing network choices. The results, expressed in terms of throughput, packet delay, and packet loss probability, outline the performance improvements with respect to other well-known networks with input buffers, such as banyan and crossbar, reached through the flexibility offered by this architectural solution.>