Cloud computing plays a crucial role in the Industry 4.0 era, particularly with the rise of Internet of Things (IoT) applications that support domains such as education, healthcare, business, and manufacturing. These applications consist of multiple services with diverse quality of service (QoS) requirements, making their development and deployment complex. While traditional cloud environments provide scalability, they often fail to support latency-sensitive and resource-intensive applications. To overcome these limitations, alternative paradigms such as Cloud–Fog–Edge (CFE), Cloud–Fog (CF), Cloud–Edge (CE), Fog–Edge (FE), and Mobile Edge Computing (MEC) have emerged. These models push computation, storage, and networking closer to end devices, reducing latency and bandwidth usage. However, the heterogeneity, mobility, and dynamic nature of these environments make service placement (a known NP-hard problem) a central challenge. This article presents a systematic literature review of service placement approaches across the compute continuum. Following established SLR methodology, we identified and analyzed 124 peer-reviewed studies published between 2018 and 2024, classifying them by (i) deployment environment, (ii) service placement strategies and algorithms, (iii) adaptability of the solution, (iv) optimization objectives, (v) virtualization/orchestration technologies, (vi) evaluation methodologies, including workloads, testbeds, and simulation tools and (vii) use cases or application types. The novelty of this work lies in providing not only a detailed taxonomy of placement approaches but also this is the first survey that takes all seven aspects into consideration and establishes correlations between them. Our findings reveal that most existing works target smart health applications and favor heuristic-based placement in complex CFE scenarios, while research on scientific and compute-intensive workloads remains limited. We also identify Kubernetes as the most widely used orchestration technology and latency as the dominant optimization metric. Despite significant progress, the field is still maturing, with gaps in real-world validation and adaptive, ML-based placement strategies. By consolidating technical approaches, evaluation practices, and open challenges, this survey offers both researchers and practitioners a structured overview of the state of the art and guidance for advancing service placement in the compute continuum.

• Proposing an FTL-based IDS framework for IoT networks using a pre-trained FL model on Bot-IoT and UNSW-NB15 datasets. • Comparing the FTL-based model with a standard FL model to assess detection performance, especially for unknown attacks. • Analyzing how varying local data distributions affect detection accuracy in federated environments. • Demonstrating improved attack detection in nodes with limited malicious data, enhancing realworld applicability. The development of Intrusion Detection Systems (IDS) for the Internet of Things (IoT) and 5G networks is rapidly advancing. This study investigates the application of federated architectures to train detection models while preserving data privacy by eliminating the need for data sharing among devices. We propose a Federated Transfer Learning (FTL) model tailored for scenarios with unbalanced nodes, enhancing the detection capabilities for unknown attacks compared to conventional Federated Learning (FL) approaches. Utilizing the Bot-IoT dataset as the source domain and the UNSW-NB15 dataset as the target domain, our experiments reveal significant improvements in detection performance. Specifically, nodes characterized by lower proportions of malicious traffic demonstrate up to a 62.614 % enhancement in detecting unknown attacks, increasing detection rates from 19.090 % to 81.704 %. Moreover, our findings indicate that FTL not only improves the identification of unknown threats but also maintains robust performance in detecting both attacks and benign traffic. Notably, the minimum accuracy achieved by the most imbalanced node reaches 0.912, in contrast to 0.741 with standard FL models. These results highlight the potential of FTL to train robust models across distributed nodes while ensuring privacy, thereby contributing to improved security measures in IoT and 5G networks.

The HORSE (Holistic, Omnipresent, Resilient Services for future 6G wireless and computing Ecosystems) framework integrates AI to address the complexity and vulnerabilities of next-generation networks. This paper focuses on an LLM-agent-based subsystem of HORSE that automates the creation and enforcement of security mitigation actions. This subsystem utilizes a specialized Knowledge Base enriched with threat-mitigation pairs from trusted sources like MITRE ATT&CK, through a collaborative workflow between an LLM and human expertise. LLMs translate high-level mitigation actions into executable Ansible commands, while a feedback-driven workflow, powered by LLM agents, ensures accuracy and reliability. One LLM agent (the "translator") generates the commands, while another (the "executor") executes them and provides feedback to the translator in case of unsuccessful execution, allowing the translator to refine the commands for a retry. Experiments with open-source LLMs (like Llama 2 and Falcon) demonstrate rates of up to 73% in execution success, and high accuracy in translating mitigation intentions. This approach automates threat mitigation, reduces human intervention, and scales effectively for the dynamic security demands of 6G networks. Future work will focus on enhancing LLM capabilities for complex environments and improving adaptability in cybersecurity operations.

This paper explores the role of data spaces in enabling secure, interoperable data exchange between Network Digital Twins and SDN Controllers. This approach emphasizes data space connectors for faster, scalable, and flexible integration of multiple components across digital ecosystems to work in unison. The proposed approach is to use a common trusted platform (in this paper, the platform used is TRUE Connector) which is responsible for secure information exchange between components(e.g: SDN Controller, Network Digital Twins, Software applications, etc.) in a digital ecosystem, thus encouraging collaboration between multiple technology vendors to provide more efficient network services and encourage interoperability.

The fifth-generation (5G) mobile network has shifted the paradigm in connectivity, high-speed data transmission, ultra-low latency, ultra-high throughput, multisense transmission, and ultra-high reliability. Today, industries are increasingly adopting 5G private networks, which also handle sensitive data related to business information, trade secrets, and personal data. Attacks on 5G private networks can potentially result in losing a competitive advantage since different security threats and vulnerabilities progressively target these networks. The unique infrastructure of the 5G network architecture and key enabling technologies exposes them to various vulnerabilities that attackers can target to breach sensitive data, steal information, and disrupt critical systems. Therefore, paying special attention to the security issues of 5G private networks is essential. The advancement of future wireless technology came about because of the different and diverse nature of connected devices, in contrast to the previous generation of mobile networks. The B5G network provides support to open network platforms, open interfaces, and the integration of different key enabling technologies helps manage network services and deploy new services needed for diverse requirements. At the same time, it has increased the attack surface compared to previous-generation networks. Therefore, it is imperative to conduct a review that focuses on addressing and classifying the different emerging threats in the private 5G and B5G networks in a distinctive way. In this paper, we have adopted a layered architecture from an industrial use case of 5G private networks to identify and classify different threats in 5G private networks. The study also characterized and modeled the different threats using information on the type of attack, entry points, and impact of the attack on the architecture layer. Moreover, an analysis of key enablers in 5G private and B5G networks and information on security threats and cyber-attacks is also presented. To accommodate the emerging threats in next-generation wireless technology, we have classified and modeled the different threats in the B5G domain using the common involved layer, which includes perception, network, and application layers in Hexa-x E2E and 6G IoT-enabled architecture. Organizations and projects that do not actively engage in cutting-edge technologies will lose their competitive advantage in the evolving technological landscape. This study has mapped the identified threat categories with different types of threats that could occur at different layers and assessed the entry points of the attacker. We have also identified the attack’s impact at each layer using security requirements related to confidentiality, availability, and integrity. Furthermore, this study reviews different AI-enabled solutions that can add value in preserving the security of 5G and B5G networks. Through this review analysis, we can conclude that the layered approach is quite beneficial in identifying the different threats and what security solutions can be offered to mitigate them.

A Digital Twin (DT) is a virtual representation of both cyber and physical objects. Initially, the concept was introduced to optimize manufacturing products, aimed at validating and verifying their behavior along their lifecycle. However, in the recent years and with the advent of Industry 4.0, it has also gained huge attention in the cybersecurity arena. Indeed, using a DT is seen as an strategy to automate, monitor, simulate, and diagnose cyber–physical systems behavior, aimed at making them to be secure. The DT can provide security in multiple domains such as manufacturing, aerospace, automotive, construction, smart grid, smart cities, and smart transportation. The fundamentals of a DT is a combination of data and a model of the system. To this end, this survey aims to explore current efforts on DT modeling techniques, and how these DT models capture the behavior of the system to enhance cybersecurity. The study focuses on the construction of behavioral models, techniques, tools, and technologies of DT used for specific security applications. The purpose of the discussion carried out in this survey is to illustrate the reader on how these behavioral models adapt to different security applications. In addition, the paper also analyzes and classifies current DT literature based on use cases, type of DT study, tools, and various DT operation modes with their specific security application. The findings of this survey describe security applications being enhanced by DT, and how the combination of key technologies facilitating Industry 4.0 (i.e., AI, IoT, Big Data and Cloud Computing, among others), are accelerating the adoption of DT as a solution for next generation security applications. Finally, this survey highlights relevant research gaps that need to be addressed before the wide adoption of DT in cybersecurity. • Provides a review of the DT modeling techniques applied to cyber security. • An in-depth analysis of recent work related to DT in enhancing security applications. • Discussion is provided on security applications, tools and architecture through DT literature analysis. • Future directions for security enhancing DT are also described.

With the increasing adoption of IoT devices and applications, significant research and development efforts have been centered around engineering novel ecosystems referred to as the IoT-edge-cloud compute continuum. In this article, we implement, analyze and present a case study for performance benchmarking of five well known and select open source MQTT broker implementations in an open-source compute continuum testbed. The proposed MQTT broker implementations are evaluated in terms of response time, different payload sizes and throughput. Measurements and results show that the hardware platform used, the message size, as well as the network parameters (latency, packet loss and jitter) have a significant impact on the resulting performance of various broker implementations and therefore have to be carefully considered in the selection process for the building blocks of the continuum. All implementations and measurements are made to be fully reproducible and free and open source.

This paper presents a novel data architecture designed specifically for smart advertising applications, focusing on the efficient flow of data while ensuring high security and data integrity. The research leverages federated learning as a black box approach, emphasizing data authenticity, privacy, and the early removal of unwanted or fake data. The proposed data model adopts a semi-random role assignment method based on various criteria to collect and aggregate data. The architecture consists of model nodes, data nodes, and validator nodes, assigned to each node based on factors such as computing power, connection quality, and historical performance. By selecting only a fraction of nodes for modeling and validation tasks, the proposed architecture optimizes resource consumption and minimizes data loss. The AROUND social network has been used as a case study to demonstrate the effectiveness of the proposed data architecture. Both, the simulation and practical implementations demonstrate that it can reduce network traffic and average CPU usage by more than 50% when the number of users increased 20 times, while maintaining the performance of the recommender model.

Chronic Respiratory Diseases (CRDs), including Chronic Obstructive Pulmonary Disease (COPD), pose significant global health challenges. Long-term oxygen therapy (LTOT) has been widely adopted to improve respiratory function and quality of life in CRD patients. However, current oxygen dosing approaches show several limitations, including inconvenience and mismatched oxygen flow rates according to patients' activity. To address these challenges, the design of Patient Specific Modeling (PSM) and the utilization of advanced technologies, such as artificial intelligence (AI), can be integrated towards a personalized oxygen dosing. This study presents a first draft of a conceptual architecture combining cloud and edge computing aimed at optimizing oxygen dosing based on individual patient data. The evaluation of sixty neural network architectures in a small-scale test has shown good results, with smaller networks generally outperforming larger ones. Further research is needed to design new testing procedures and explore optimal parameters for the neural networks, including a large patients' number.

Smart advertising is growing in popularity and affecting businesses. Smart advertising is a more friendly, inter-active, personalised, and creative method of promoting a product, and attempts to delight clients. AROUND is a social networking service that emphasizes smart advertising through an effective recommender system. The system considers user profiles, history, social network connections, mood, and IoT-supported positioning to select the most relevant ads using machine learning technology. Although the current deployment of the AROUND system is based on the cloud, an edge-based architecture provides relevant improvement in terms of system response time. In this paper we extend the edge-based strategy to leverage the potential of federated learning on multiple distributed edge servers. We show that federated learning can take advantage of the distributed nature of the system, and leverage the specificities of local features. In fact, in this research, we propose a novel federated learning solution to provide smart advertising as a classification problem which uses ensemble methods and logistic regression as internal (local) models and meta-heuristic algorithms for federated learning aggregation. As part of the experiments, we prove this technology on a real data set with more than one million registers, and show the efficiency in terms of enhanced accuracy and improved training and response speed.

In recent years, the number of embedded computing devices connected to the Internet has exponentially increased. At the same time, new applications are becoming more complex and computationally demanding, which can be a problem for devices, especially when they are battery powered. In this context, the concepts of computation offloading and edge computing, which allow applications to be fully or partially offloaded and executed on servers close to the devices in the network, have arisen and received increasing attention. Then, the design of algorithms to make the decision of which applications or tasks should be offloaded, and where to execute them, is crucial. One of the options that has been gaining momentum lately is the use of Reinforcement Learning (RL) and, in particular, Deep Reinforcement Learning (DRL), which enables learning optimal or near-optimal offloading policies adapted to each particular scenario. Although the use of RL techniques to solve the computation offloading problem in edge systems has been covered by some surveys, it has been done in a limited way. For example, some surveys have analysed the use of RL to solve various networking problems, with computation offloading being one of them, but not the primary focus. Other surveys, on the other hand, have reviewed techniques to solve the computation offloading problem, being RL just one of the approaches considered. To the best of our knowledge, this is the first survey that specifically focuses on the use of RL and DRL techniques for computation offloading in edge computing system. We present a comprehensive and detailed survey, where we analyse and classify the research papers in terms of use cases, network and edge computing architectures, objectives, RL algorithms, decision-making approaches, and time-varying characteristics considered in the analysed scenarios. In particular, we include a series of tables to help researchers identify relevant papers based on specific features, and analyse which scenarios and techniques are most frequently considered in the literature. Finally, this survey identifies a number of research challenges, future directions and areas for further study.

Fog computing is a recent paradigm leveraging available resources at the edge of the network intended to extend the traditional cloud model towards the novel cloud continuum computing model. Recognized the unstoppable growth of highly dynamic and heterogeneous edge devices, as well as the pop up of a large set of diverse and ever more demanding services, the selection of those edge resources best meeting service requirements while also matching the expected QoS guarantees is, with no doubt, a challenging task. This paper presents a rank-based model aimed at both evaluating edge nodes’ characteristics and selecting nodes best performing the controller role, whilst simultaneously satisfying the required QoS constraints, coining the so-called Control-as-a-Service concept. To that end, a yet simple prediction strategy, based on Dynamic Branch Prediction is introduced to avoid unnecessary controller exchanges and QoS degradation. In the performed experiments, the proposed method yielded a reduction in the number of exchanges when compared to a solution with no prediction, under different scenarios. Comparing distinct selection strategies, the proposed model presented an improvement in controller availability as well as in relevant controllers’ characteristics, such as battery and memory capacity.

Supply chains constitute a type of IoT system that is becoming increasingly complex today. Providing supply chains with cybersecurity guarantees is becoming a highly topic in current research. This interest is mainly rooted on the fact that supply chains may include different vulnerable components, such as those residing in factories, distribution centres, logistics, or transportation, just to name a few. Moreover, most of these components are controlled by different IoT devices, such as sensors and actuators, also becoming well-known sources of vulnerabilities, easing attackers access to the whole system. In this scenario, cyberattacks that use this weak link in a supply chain are increasing in the last years, as it has been largely reported in the literature. In this heterogeneous scenario, the key trend to facilitate cybersecurity provisioning sits on guaranteeing cyber-security in the entire supply chain, handled as a whole, that is, not separately for each individual component of the chain. In this paper, we propose an authentication/access mechanism for the weaker elements in a supply chain, i.e., the IoT devices, leveraging blockchain as the baseline technology, based on considering the generation of temporal tickets to have a higher access control.

Smart advertising creates awareness about some offer with a more direct, personalized and interactive focus. In this area, AROUND is a social network aimed at providing smart advertising to suggest appealing business to their customers and friends. The AROUND system is supported by a sophisticated recommender system, which considers not only the customers historical behaviours, but also their current mood and accurate location. In such smart recommendation systems, the response time for the personalized advertising is critical for a successful users’ quality of experience. In this research work we first evaluate the current performance of the AROUND system in terms of processing and communication times considering that, nowadays, this social network has more than 3 million users. The current implementation of the system relies on the deployment of a network of beacons, and uses a domestic cloud provider as the main infrastructure. We show that when the number of concurrent requests becomes too high, the response time faces some limitations. In order to address this issue, we discuss several alternatives, and propose the use of an edge-based strategy as a solution for fast response time. In the experimental section, we measure the performance of the AROUND system, both in our current infrastructure at the cloud and with an edge-based approach, and show the additional advantages of leveraging the edge-based strategy even in the case of overloading the cloud capacity.

In recent years, computer network architectures are experiencing a significant shift motivated by a myriad of edge devices generating a tremendous volume of data, service providers deploying real-time and huge bandwidth-consuming network applications, and mobile end-users demanding stringent quality of service and reduced service disruption. The fog computing architecture aims at addressing several related issues by employing computing resources at the edge of the network. However, frequent and even unexpected handover among distinct fog domains is yet a research challenge because it hinders the continuous availability of shared edge resources. In this work, we employ reinforcement learning (RL) to learn from experience how to maximize the availability of resources at fog domains by minimizing the handover frequency in vehicular scenarios through smart resource placement. We evaluated our RL-based model in simulations mimicking real-world scenarios where each moving vehicle may connect to different fog domains throughout its route. The results show that the proposed model yields an improvement in the availability of resources in comparison to a greedy strategy under all simulated scenarios.

Smart cities are becoming popular and, with them, a plethora of resources is emerging turning into a massive concentration of computing devices. In such environments, the deployment of a smart resources management system is envisioned as an extraordinary opportunity to provide almost unlimited computing and networking capabilities obtained by putting together a vast set of nearby devices, hence offering high performance computing while benefiting from low latency networking. In this scenario, where millions of highly heterogeneous devices are available, a key challenge is to select the set of devices suited the most to carry out some specific tasks. In this paper, we present a performance estimation model intended to find an optimal solution for load distribution over massive heterogeneous systems, considering the effects of both, distributed computing and network overhead. The current implementation is in a preliminary stage, considering only fully parallel applications. Trials conducted considering several hardware configurations, with either homogeneous or heterogeneous systems, show the fact that the predicted performance matches the one obtained by the experimental executions with an almost negligible deviation. These initial results encourage us to continue developing the model for more complex and realistic applications.

There are a myriad of computing resources in the World, most of which are underused. Efficiently using some of these devices raises several interesting opportunities. For instance, devices are spread everywhere, so locality can be deeply exploited. Furthermore, combining the computational capacity of all, or some, of them may conform a virtual cluster more powerful than any current data center. And, finally, by using resources already there avoids the need for a specific investment on specific infrastructures, as well as significantly reduces the operational costs. In this paper we propose a research architecture for global resources management which aims at exploiting the availability of this vast amount of heterogeneous resources at the edge of the network. We have designed the Progressive Mapping System (PMS) as a system that keeps track of the available resources and, upon an application launching, finds the best devices configuration for optimal application execution, considering both, the application performance requirements and the resources availability and performance features, while guaranteeing the Quality of Service (QoS). The PMS implements a graph of devices where nodes describe their performance features and edges describe their geographic distance as well as communication capabilities. Given an application description, which includes some performance description and quality of service requirements, the PMS system selects a subset of the global graph by solving a mixed integer quadratically constrained programming formulation, which finds the appropriate set of nodes for optimal application execution performance. One of the novelties of the PMS system is that it considers the geographic positioning of the devices and the application locality requirements to deeper exploiting locality issues. The PMS system is yet in a preliminary implementation stage. In this paper we describe the system overview, the architectural design, and discuss some research challenges which are critical for an effective PMS implementation.

The inherent mobility characterizing users in fog computing environments along with the limited wireless range of their serving fog nodes (FNs) drives the need for designing efficient mobility management (MM) mechanisms. This ensures that users' resource-intensive tasks are always served by the most suitable FNs in their vicinity. However, since MM decisionmaking requires control information which is difficult to predict accurately a-priori, such as the users' mobility patterns and the dynamics of the FNs, researchers have started to shift their attention towards MM solutions based on online learning. Motivated by this approach, in this paper, we consider a bandit learning model to address the mobility-induced FN selection problem, with a particular focus on scenarios with a high FN density. Following this approach, a software agent implemented within the user's device learns the FNs' delay performances via trial and error, by sending them the user's computation tasks and observing the perceived delay, with the goal of minimizing the accumulated delay. This task is particularly challenging when considering a high FN density, since the number of unknown FNs that need to be explored is high, while the time that can be spent on learning their performances is limited, given the user's mobility. Therefore, to address this issue, we propose to limit the number of explorations to a small subset of the FNs. As a result, the user can still have time to be served by the FN that was found to yield the lowest delay performance. Using real world mobility traces and task generation patterns, we found that it pays off to limit the number of explorations in high FN density scenarios. This is shown through significant improvements in the cumulative regret as well as the instantaneous delay, compared to the case where all newly-appeared FNs are explored.

In the recent years, fog computing has been proposed as a promising paradigm to enhance the performance of latency-critical applications by hosting them within idle computation nodes (Fog Nodes, FNs) at the edge of the network. Such a paradigm has been further extended to the vehicular fog computing (VFC) scenario by leveraging available computing resources offered by modern vehicles, thus allowing them to process different types of computation tasks on behalf of onboard users or nearby vehicles. Within this context, the proper assignment of computation tasks to Vehicular FNs is an important issue that is currently under active research. To address this issue, online learning approaches, where the performances of the Vehicular FNs in terms of task execution delay are learnt via trial and error, are starting to gain in popularity. This is mainly motivated by the inherent uncertainty caused by mobility and fluctuating resource availabilities in VFC environments. However, since the process of learning from scratch in such a dynamic vehicular environment may lead to a degradation in the learning performance, this paper proposes the use of an advising mechanism, where a roadside unit (RSU) who has already learnt the performances of the vehicles within its range, uses its acquired knowledge to provide advice to a neighbor RSU who does not have enough experience allowing it to make efficient assignment decisions on its own. To evaluate this approach, we used realistic vehicular mobility traces to simulate the VFC scenario. The obtained results show that our proposed approach improves the learning performance compared to the case where no advice is leveraged.

A vast amount of data is constantly being produced in the world from several sources, including the IoT at smart cities. Such complex universe of digital data is being set up through these daily accumulated fresh data over other historical repositories. This data can then be used in different forms and then be reused in further processes, hence, drawing the life cycle of data. Conventionally, the smart city resources management are based on cloud models where sensors data are collected to prepare a centralized and rich set of open data. Cloud-based frameworks provide several benefits, consisting of their ubiquity and (almost) unlimited resources capacity. But accessing data from the cloud results in large network traffics, high latencies and are not recommended for real-time solutions. In order to solve this, fog computing has been defined as a promising technology. It uses devices at the edge to perform closer computation and, therefore, reduces network traffic, decreases latencies, while enhancing security levels. In this paper we present the Data Preservation block, as part of our hierarchical fog-to-cloud F2C data management architecture. This model has the advantages of both fog and cloud technologies, as it allows reduced latencies for critical applications while being able to use the high computing capabilities of cloud technology. Furthermore, we have estimated the amount of collected data to be stored in the storage media, from fog to cloud layers, at the Barcelona smart city, and shown the tremendous advantages of such data organization.

In order to fully leverage idle computational resources in a fog computing scenario, a suitable discovery mechanism is needed. Within this context, F2C-Aware has been proposed in a previous work as a discovery solution where a device can detect nearby fog resources by scanning for custom 802.11 beacons being broadcast by those resources. Since mobility is an inherent characteristic in fog computing, keeping the Wi-Fi-based discovery service always on during mobility will lead to unnecessary scans being performed in non-covered areas, causing energy consumption penalties. That is why, in this paper, we propose a Beacon-assisted Direction-aware Scanning Scheme (BDSS), where we use the discovery beacon to convey the distance remaining until a next fog will be reached in each of the four cardinal directions along with the channel used for beacon advertisements. Therefore, the scan will be disabled until the next fog is reached. Emulation results show that by using this scheme, considerable energy savings could be made while maintaining a high discovery rate.

Recently, new computing paradigms such as fog and edge computing started to emerge in an attempt to cope with the low latency requirements of new classes of IoT applications. In order for these paradigms to fully realize their potential, an important challenge to address is the discovery of fog nodes (FNs) with spare computational resources that can be used to host time- sensitive and computationally intensive application components. We particularly focus on this problem in this paper, considering the case of a WiFi-based FN discovery process. More specifically, we evaluate how practical it is to trigger the discovery of fog nodes based on a mobile phone's historical cellular footprints in order to obtain a high discovery rate and a low energy overhead. To this end, we conducted a small-scale cellular data collection to be used to test different learning approaches, including the K-Nearest Neighbors and the decision tree algorithms as well as a Hidden Markov Model (HMM). According to our evaluation results, HMM was found to achieve the maximum discovery and energy saving ratios. The impact of initial FN misdetections on the user-FN contact ratio has also been studied.

Fog-to-Cloud (F2C) is an emerging architecture intended to manage the resources continuum from far datacenters up to the near edge, putting together the cloud and fog concepts. It aims to obtain both, an efficient utilization of the entire set of available resources and an optimal execution of highly demanding services, supported by the deployment of highly skilled IoT devices at the edge. Many research efforts are pushing for the creation of a widely accepted F2C framework architecture aimed at controlling and managing the resources integrating the F2C network in a hierarchical and distributed fashion. To that end however, many research challenges remain unsolved. This paper puts the focus on one of these challenges, particularly the identity management and proposes a novel strategy that, based on fragmenting the full resource identifier, uses portions of the full resource name to identify the resources in the network without losing the uniqueness property. The presented results show that by using fragments instead of the full identifier, the lookup time and database size in the F2C aggregator nodes are both notably reduced, what undoubtedly enables a fast resource identification.

The wide spread deployment of smart edge devices and applications that require real-time data processing, have with no doubt created the need to extend the reach of cloud computing to the edge, recently also referred to as Fog or Edge Computing. Fog computing implements the idea of extending the cloud where the

Traditional smart city resources management rely on cloud based solutions to provide a centralized and rich set of open data. The advantages of cloud based frameworks are their ubiquity, (almost) unlimited resources capacity, cost efficiency, as well as elasticity. However, accessing data from the cloud implies large network traffic, high data latencies, and higher security risks. Alternatively, fog computing emerges as a promising technology to absorb these inconveniences. The use of devices at the edge provides closer computing facilities, reduces network traffic and latencies, and improves security. We have defined a new framework for data management in the context of smart city through a global fog to cloud management architecture; in this paper we present the data acquisition block. As a first experiment we estimate the network traffic during data collection, and compare it with a traditional real system. We also show the effectiveness of some basic data aggregation techniques in the model, such as redundant data elimination and data compression.

The recent deployment of novel network concepts, such as M2M communication or IoT, has undoubtedly stimulated the placement of a new set of services, leveraging both centralized resources in Cloud Data Centers and distributed resources shared by devices at the edge of the network. Moreover, Fog Computing has been recently proposed having as one of its main assets the reduction of service response time, further enabling the deployment of real-time services. Albeit QoS-aware network researches have been originally focused on data plane issues, the successful deployment of real-time services, demanding very low delay on the allocation of distributed resources, depends on the assessment of the impact of controlling decisions on QoS. Recently, Fog-to-Cloud (F2C) computing has been proposed as a hierarchical layered-architecture relying on a coordinated and distributed management of both Fog and Cloud resources, enabling the distributed and parallel allocation of resources at distinct layers, thus suitably mapping services demands into resources availability. In this paper, we assess the layered management architecture in F2C systems, taking into account its distributed nature. Preliminary results show the tradeoff observed regarding controllers capacity, number of controllers, and number of controller layers in the F2C architecture.

A huge amount of data is constantly being produced in the world. Data coming from the IoT, from scientific simulations, or from any other field of the eScience, are accumulated over historical data sets and set up the seed for future Big Data processing, with the final goal to generate added value and discover knowledge. In such computing processes, data are the main resource; however, organizing and managing data during their entire life cycle becomes a complex research topic. As part of this, Data LifeCycle (DLC) models have been proposed to efficiently organize large and complex data sets, from creation to consumption, in any field, and any scale, for an effective data usage and big data exploitation.

There is a vast amount of data being generated every day in the world, coming from a variety of sources, with different formats, quality levels, etc. This new data, together with the archived historical data, constitute the seed for future knowledge discovery and value generation in several fields of eScience. Discovering value from data is a complex computing process where data is the key resource, not only during its processing, but also during its entire life cycle. However, there is still a huge concern about how to organize and manage this data in all fields, and at all scales, for efficient usage and exploitation during all data life cycles. Although several specific Data LifeCycle (DLC) models have been recently defined for particular scenarios, we argue that there is no global and comprehensive DLC framework to be widely used in different fields. For this reason, in this paper we present and describe a comprehensive scenario agnostic Data LifeCycle (COSA-DLC) model successfully addressing all challenges included in the 6Vs, namely Value, Volume, Variety, Velocity, Variability and Veracity, not tailored to any specific environment, but easy to be adapted to fit the requirements of any particular field. We conclude that a comprehensive scenario agnostic DLC model provides several advantages, such as facilitating global data organization and integration, easing the adaptation to any kind of scenario, guaranteeing good quality data levels, and helping save design time and efforts for the research and industrial communities.

The novel Fog-to-Cloud (F2C) computing paradigm has been recently proposed aiming at the enhanced integration of Fog Computing and Cloud Computing through the coordinated management of underlying resources, taking into account the peculiarities inherent to each computing model, and enabling the parallel and distributed execution of services into distinct fog/cloud resources. Nevertheless, studies on F2C are still premature and several issues remain unsolved yet. For instance, in an F2C scenario service allocation must cope with the specific aspects associated to cloud and fog resource models, requiring distinct strategies to properly map IoT services into the most suitable available resources. In this paper, we propose a QoS-aware service distribution strategy contemplating both service requirements and resource offerings. We model the service allocation problem as a multidimensional knapsack problem (MKP) aiming at an optimal service allocation taking into consideration delay, load balancing and energy consumption. The presented results, demonstrate that the adopted strategy may be applied by F2C computing reducing the service allocation delay, while also diminishing load and energy consumption on cloud and fog resources.

The recent technological advances related to computing, storage, cloud, networking and the unstoppable deployment of end-user devices, are all coining the so-called Internet of Things (IoT). IoT embraces a wide set of heterogeneous services in highly impacting societal sectors, such as Healthcare, Smart Transportation or Media delivery, all of them posing a diverse set of requirements, including real time response, low latency, or high capacity. In order to properly address such diverse set of requirements, the combined use of Cloud and Fog computing turns up as an emerging trend. Indeed, Fog provides low delay for services demanding real time response, constrained to support low capacity queries, whereas Cloud provides high capacity at the cost of a higher latency. It is with no doubt that a new strategy is required to ease the combined operation of cloud and fog infrastructures in IoT scenarios, also referred to as Combined Fog-Cloud (CFC), in terms of service execution performance metrics. To that end, in this paper, we introduce and formulate the QoS-aware service allocation problem for CFC architectures as an integer optimization problem, whose solution minimizes the latency experienced by the services while guaranteeing the fulfillment of the capacity requirements.

The recent advances in optical technologies pave the way to the deployment of high-bandwidth services. As reliability becomes a mandatory requirement for some of these services, network providers must endow their networks with resilience capabilities. In recent years, network coding protection (NCP) has emerged as a tentative solution aiming at enabling network resilience in a proactive and efficient way. The goal of this paper is to conduct a techno-economic study to evaluate the protection cost required by NCP schemes deployed either at the IP/MPLS or at the Optical layer of a multi-layer network, as well as its impact on both the capital and operational expenditures (CAPEX, OPEX) of a network provider. Our evaluation results show that a significant reduction in both CAPEX and OPEX is obtained with NCP. Indeed, at least a 49% and 52% of CAPEX and OPEX reduction is achieved respectively in comparison with conventional proactive protection schemes.

A route leak can be defined as a security gap that occurs due to the infringement of the routing policies that any two Autonomous Systems (ASes) have agreed upon. Route leaks are seemingly simple, but hard to resolve since the ASes keep their routing policies confidential. Indeed, the traditional palliatives, such as the utilization of route filters, are no longer used by a large number of ASes, given the high administrative burden that they entail. Other alternatives, like BGP monitoring tools, not only require third party information gathered at multiple vantage points, but also they become impotent in many cases, due to their limited view of the interdomain routing state. In this paper, we propose a different approach, which allows to autonomously detect the occurrence of route leaks by solely inspecting the BGP information available at the AS. Our main contributions can be summarized as follows. First, we propose a self-contained Route Leak Detection (RLD) technique, which is based on real-time analytics on the Route Information Bases (RIBs) of the border routers of an AS. Second, we introduce Benign Fool Back (BFB), "a harmless bluff" that can substantially improve the success rate of the RLD technique. Third, we show through exhaustive simulations that our technique can detect route leak incidents in various scenarios with high success rate. In addition, our solution has the following practical advantages: a) no reliance on third party information (e.g., on vantage points); b) no changes required to control-plane protocols (e.g., to BGP); and c) allows non-invasive integration (e.g., using SDN).

The motivation behind the Locator/Identifier Separation Protocol (LISP) has shifted over time from routing scalability issues in the core Internet to a set of use cases for which LISP stands as a technology enabler. Among these are the mobility of physical and virtual appliances without breaking their TCP connections, seamless migration and fast deployments of IPv6, multihoming, and data-center applications. However, LISP was born without security, and therefore is susceptible to attacks in its control-plane. The IETF's LISP working group has recently started to work in this direction, but the protocol still lacks end-to-end mechanisms for securing the overall registration process on the mapping system. In this paper, we address this issue and propose a solution that counters the attacks. We have deployed LISP in a real testbed, and compared the performance of our proposal with current LISP implementations, in terms of both messaging and packet size overhead. Our preliminary results prove that our solution offers much higher security with minimum overhead.

Recent research in the field of Wireless Sensor Networks (WSNs) has demonstrated the advantages of using learning automata theory to steer the routing decisions made by the sensors in the network. These advantages include aspects such as energy saving, energy balancing, increased lifetime, the selection of relatively short paths, as well as combinations of these and other goals. In this paper, we propose a very simple yet effective technique, which can be easily combined with a learning automaton to dramatically improve the performance of the routing process obtained with the latter. As a proof-of-concept, we focus on a typical learning automata-based routing process, which aims at finding a good trade off between the energy consumed and the number of hops along the paths chosen. In order to assess the performance of this routing process, we apply it on a WSN scenario where a station S gathers data from the sensors. In this typical WSN setting, we show that our combined technique can significantly improve the decisions made with the automata; and more importantly, even though the proof-of-concept particularizes somehow the automata and their behavior, the technique described in this paper is general in scope, and therefore can be applied under different routing methods and settings using learning automata.

The Path Computation Element (PCE) is positioned nowadays as one of the solutions that almost every carrier will eventually deploy. The PCE architecture as well as a number of components, including the PCE Communication Protocol (PCEP), have been standardized by the IETF. However, a number of challenges remain to be solved on its way from standards to deployment. In particular, the existing proposals for multilayer path computation within the PCE framework need to be further developed and tested, before considering their possible integration into operational networks. This is especially true for the interoperability of the various PCE implementations and the extensions such as the Virtual Network Topology Manager (VNTM) which cannot be taken for granted. This paper presents a functional implementation of coordinated computation of multilayer paths supported through inter-layer PCE communication, where one PCE is developed by industry and the other as an open-source effort. To this end, we consider an IP/MPLS network deployed over a Wavelength Switched Optical Network (WSON), each of which deploying its own PCE, in an attempt to create an inter-operable multilayer solution. We discuss the key challenges that the research community will face in this area, which in turn will drive a considerable part of the upcoming efforts in terms of standardization.

Traditional network architectures based on the Internet Protocol (IP) are now being questioned by the research community, since they are no longer positioned as the most suitable paradigm for supporting the increasing diversity of applications and uses of the Internet. A key issue in this subject is that, although the IP protocol has provided the basis for the rapid evolution of the Internet, its addressing scheme is not prepared to face the challenges posed by many foreseen applications. In light of this, different initiatives worldwide have started specific research programs to address these problems and work toward the "Future Internet". The TARIFA project represents one of these initiatives, and it is positioned as a clean slate alternative aimed at overcoming the critical issues in today's Internet. The novelty in TARIFA resides in the fact that any “commodity” in the network can be composed as a set of atomic services, which can be in turn assembled through a service-centric model for building a promising Internet architecture. In this paper, we focus on the space requirements and set the basis for a new addressing scheme suitable for service-centric network architectures such as the one proposed by TARIFA. The addressing scheme discussed in this paper is general in scope, and could be applied not only to architectures based on the composition of services but also to user and data-centric Internet architectures.

The success of Peer-to-Peer (P2P) overlay networks has led to its broad adoption on different scenarios, thereby increasing the set of features present on such protocols, e.g., audio and video-conferencing, streaming of multimedia content using Peer-to-Peer TeleVision (P2PTV), and so on. The wide acceptance of P2P multimedia streaming by the end-users has captured the attention of content providers, making it a suitable candidate to be adopted as technology for the offered multimedia services. Nevertheless, to deliver a proper service, content providers, need to quantify and assess the quality perceived by the users. In the past, this assessment was embedded into Service Level Agreements under the client/server paradigm, but with the introduction of P2P, these solutions became unsuitable. To overcome this limitation, we present a Monitoring and Management Framework to assess the Quality of Service and the Quality of Experience of multimedia traffic in any P2PTV streaming application. The proposed approach is designed to potentially assist management systems to actively assess the performance of existing streaming deployments. To demonstrate the usefulness of the system, we, as an use-case, deploy it in a real scenario where we analyze the performance of a particular streaming application in a P2PTV environment.

The rapid growth of network services is forcing operators to use of inter-layer coordination between the IP and the transport network management systems in order to optimize resource utilization within the carriers network ecosystem. However, the diversity of technology and the related management systems within the IP and transport networks poses significant challenges in developing a single unified management ecosystem for both IP and transport networks. In this work, we present a basic prototype of a Service Oriented Architecture (SOA) based framework proposed in the EU project ONE for programmable orchestration of IP and transport management processes. The proof-of-concept prototype uses web services and BPEL to facilitate programmable inter-layer coordination, and was tested on a number of use cases. We measured the performance penalty of using the SOA framework against the traditional JAVA based provisioning approaches used in integrated multi-layer management, and show that the penalty incurred in execution times were sufficiently small while bringing several significant advantages during the software design, development, integration and maintenance, which is simple and highly effective.

This paper presents a Cross-layer Network Management System (NMS) that allows Service Providers (SPs) to perform cost-effective network resource reservations with their Network Operators (NOs). The novelty of our NMS is that it offers a fresh and promising approach to use the end-users satisfaction level as the metric to perform the resource management. We show that our system is capable of achieving considerable reductions in the operational costs of SPs, while keeping proper bounds in the end-user satisfaction for the offered multimedia services.

The Locator/Identifier Separation Protocol (LISP) is one of the candidate solutions to address the scalability issues in inter-domain routing. The current proposals for its control plane (e.g., ALT, CONS, NERD) have various shortcomings, including the potential dropping of packets at LISP routers during the resolution of the EID-to-RLOC mapping. In this paper, we introduce a new Control Plane (CP) for LISP supported by an architecture that borrows concepts from both the Path Computation Element (PCE) and Intelligent Route Control (IRC). Our CP is able to tackle three different problems simultaneously: (i) packets sourced from end-hosts are neither dropped nor queued during the mapping resolution; (ii) the EID-to-RLOC mapping can be obtained and configured approximately within the DNS resolution time needed to fetch the destination EID address; and (iii) our approach can blend IRC with the PCE capabilities, to perform upstream/downstream Traffic Engineering (TE) through the dynamic management of the mappings. In particular, our CP supports the utilization of different LISP ingress and egress local routers for the same flow sourced from a domain.

At present, service providers have several incentives to extend the reach of long-lived MPLS paths across domains. Providers, however, will face a number of trade-offs while choosing the optimal set of MPLS paths to be established. In this paper, we focus on the multi-objective decision problem of maximizing the traffic demands to be covered by long-lived MPLS paths from a source domain S to its major destination domains, while minimizing the monetary costs incurred. The problem is formulated subject to a budget constraint, which assures the minimum expected revenue for the provider in S. A major advantage of the analysis and solution proposed in this paper is that it can be easily generalized, and applied in other settings where constrained problems considering maximum coverage vs. cost are critical.

We present a distributed routing algorithm for finding two disjoint (primary and backup) QoS paths that run across multiple domains. Our work is inspired by the recent interest in establishing communication paths with QoS constrains spanning multiple IP/MPLS domains. In such settings, the routing decisions in each domain are made by the path computation element (PCE). We assume that the PCEs run a joint distributed routing protocol, decoupled from the BGP, which enables them to establish efficient paths across multiple domains. This study makes the following contributions. First, we present an aggregated representation of a multi-domain network that is small enough to minimize the link-state overhead, and, at the same time, is sufficiently accurate, so that the PCEs can find optimal disjoint QoS paths across multiple domains. Second, we present a distributed routing algorithm that uses the proposed representation to find disjoint paths in an efficient manner. Finally, we consider the problem of finding two disjoint paths subject to the export policy limitations, imposed by customer-provider and peer relationships between routing domains. We show that this problem can be efficiently solved by employing the concept of line graphs. To the best of our knowledge, this is the first scheme fully decoupled from BGP that enables to establish disjoint QoS IP/MPLS paths in a multi-domain environment with provable performance guarantees.

This paper presents a RWA strategy based on the stochastic estimation of the effective number of available wavelengths (ENAW) along interdomain paths. We propose an approximate model to roughly estimate the ENAW on the paths across multiple domains, and then refine this estimation by means of observations and an adaptive prediction-correction Kalman filtering process. Unlike traditional Kalman filters, which use noisy measurements as their observations, we use the information contained in the routing updates as noisy measurements of the existent wavelength occupancy along the paths. Based on these observations, we estimate the ENAW on the candidate paths, and use this information to influence the RWA decision when the number of available wavelengths makes a lightpath request prone to be blocked. The approximate model we are proposing in this paper is based on a noisy extension of a simplified model derived for two domains. Our results validate the usefulness of the model, and confirm that by estimating the wavelength occupancy prior to the RWA decision the blocking ratio can be considerably reduced.

Usual QoS routing algorithms involve the periodic update of network state information in all the network nodes. Based on this knowledge the QoS routing algorithms select the `best' route. It has been shown in the literature that the performance of these QoS routing algorithms strongly depends on the frequency of updating. We propose a new QoS routing mechanism called Prediction-Based Routing based on predicting the availability of links and routes regardless from the network state information. Consequently, update messages are not required, hence reducing signalling overhead and providing a major enhancement in terms of scalability. We show that the PBR is a viable option compared with usual QoS routing algorithms from the point of view of performance, complexity ad signalling overhead.

Smart route control is being increasingly used as a way to dynamically improve the end-to-end performance of the outbound traffic of multihomed stub domains. However, all the solutions available at present have in common two drawbacks which are key motivations for this work. First, all solutions are standalone, so no routing control interactions exist between the domains sourcing and sinking the traffic. The consequences of this lack of interactions are quite coarse route control over the outbound traffic from the domains, and the inability to smartly control how traffic flows into the domains. The second drawback is that all available solutions behave in a fully selfish way, that is, they operate without considering the effects of their decisions in the performance of the network. Given these limitations, we propose to extend the existing route control model from standalone and selfish to a cooperative and social route control model. Our main contribution in this paper is to show that when several route controllers compete for the same network resources, the conventional ones are outperformed by those using a cooperative and social approach and this becomes especially noticeable as the network utilization increases. Our results reveal that it is possible to reduce the frequency of traffic relocations by more than a 50% on average and still obtain slightly better end-to-end traffic performance for delay-sensitive applications. A key advantage is that our extension can be installed and used today by simply performing software upgrades to any of the existing route control solutions.

A hierarchical network architecture is one of the hard recommendations stated at the automatically switched optical networks (ASON) specifications. This paper focuses on providing ASON with a hierarchical routing in order to ensure the scalability for large worldwide networks, mainly focusing on the functionality expected from GMPLS routing, such as aggregation schemes and routing algorithms, targeting to optimize the global network performance while guaranteeing scalability. Simulation results indeed show the benefits obtained owing to harder overhead reduction without impacting on the blocking probability.

In large dynamic networks it is extremely difficult to maintain accurate routing information on all network nodes. Different causes can motivate this inaccuracy, such as the state aggregation produced in hierarchical networks, the delay in flooding the network state, and the triggering policy used to determine when this network state information must be updated. This paper focuses on the inaccuracy caused by the triggering policies. Triggering policies are included in the routing protocol to reduce the large number of update messages needed to guarantee accurate network state information on all the network nodes. The BYPASS based optical routing (BBOR) has already been proposed by the authors to reduce the effects of having inaccurate routing information in networks operating under the wavelength-continuity constraint This paper extends the BBOR mechanism to be applied to wavelength convertible networks and evaluates its performance.

Multiprotocol label switching and traffic engineering are proposed by the IETF to improve the network performance. Moreover, some QoS routing algorithms must be added in order to optimize the path selection process. However, in highly dynamic large networks, existing QoS routing algorithms suffer from a blocking probability, which in part is due to the existence of inaccuracy in the network state information used to select the path. This paper deals with the BYPASS based routing mechanism, which was introduced in the previous paper to overcome this routing inaccuracy effect in IP/MPLS scenarios. We suggest an enhancement of the BBR mechanism to optimize the bandwidth allocation by balancing the path length and the residual bandwidth, and its performance is evaluated by simulation.

Multiprotocol label switching (MPLS) is an advanced forwarding scheme, which allows the network to achieve the traffic engineering (TE) objectives. When two or more MPLS nodes are connected via an ATM backbone merely composed of ATM switches, it is necessary to add some mechanisms to allow the label distribution and allocation between both nodes. In this paper a solution is proposed to integrate ATM and MPLS, so that the ATM backbone will be seen as a unique MPLS node from the MPLS domain and as a consequence of that the end-to-end LSP establishment can be carried out.