The emergence of the fifth generation (5G) technology has transformed mobile networks into multi-service environments, necessitating efficient network slicing to meet diverse Service Level Agreements (SLAs). SLA decomposition across multiple network domains, each potentially managed by different service providers, poses a significant challenge due to limited visibility into real-time underlying domain conditions. This paper introduces Risk-Aware Iterated Local Search (RAILS), a novel risk model-driven meta-heuristic framework designed to jointly address SLA decomposition and service provider selection in multi-domain networks. By integrating online neural network (NN)-based risk modeling with iterated local search principles, RAILS effectively navigates the complex optimization landscape, utilizing historical feedback from domain controllers. We formulate the joint problem as a Mixed-Integer Nonlinear Programming (MINLP) problem and prove its NP-hardness. Extensive simulations demonstrate that RAILS achieves near-optimal performance, offering an efficient, real-time solution for adaptive SLA management in modern multi-domain networks.

In-band network telemetry (INT), enabled by programmable data planes and the appearance of programming protocol-independent languages such as P4, emerged as a viable approach for network monitoring. INT allows the collection of fine-grained network information in real-time, increasing network visibility, at the cost of network overhead. Several lightweight INT approaches have been recently proposed that attempt to alleviate the transmission overhead of INT, while maintaining a high degree of monitoring accuracy. However, their impact on the resources of the respective hardware network devices has been hardly investigated as most of the approaches are evaluated via simulation. In this study, we provide proof of concept implementations of two lightweight INT approaches that have been proposed for path tracing on the Intel Tofino ASIC, identifying the challenges of porting the solution to the selected target. We examine their performance, providing an in-depth analysis of resource consumption.

In the context of 6G, Digital Twin technology has the potential to play a crucial role by offering a sophisticated, dynamic model that mirrors the physical world in real-time. Integrating digital twins into 6G enables precise orchestration and optimization of network resources, meeting the diverse and stringent demands of next-generation applications. Furthermore, digital twins provide a platform for continuous learning and AI-driven analysis. Research on the architecture of Network Digital Twins (NDTs) for 5G/6G is still limited, often focusing on partial implementations rather than comprehensive, full-stack approaches. This paper proposes a high-level architectural framework for 6G NDTs system, structured across three layers: the Physical Twin Layer, Digital Twin Layer, and Application and Service Layer. We discuss the challenges of deploying these systems and the importance of selecting appropriate tools, presenting an experimental case study that demonstrates the impact of different tool choices on system performance. Our findings underscore the need for flexible, scalable solutions to fully realize the benefits of NDTs in 6G networks.

To maintain the required service quality of time-critical cloud applications, operators must continuously monitor their runtime status, detect potential performance anomalies, and diagnose the root causes of these anomalies effectively. However, existing performance diagnosis methods face challenges such as the need for high-quality labeled data, the low reusability and robustness of performance anomaly detection models, and the absence of real-time fine-grained root cause localization. These challenges make fixing performance issues quickly and developing effective adaptation decisions difficult. We provide a Fine-grained Robust Performance Diagnosis (FIRED) framework to tackle those challenges. The framework offers a metrics selection component to filter noise and improve detection efficiency, an anomaly detection component that assembles several well-selected base models with a deep neural network, and adopts weakly supervised learning considering fewer labels exist in reality. The framework also employs a real-time, fine-grained root cause localization component to locate dependent resource metrics of performance anomalies. Our experiments show that the framework can effectively reduce data noise and achieve the best accuracy and algorithm robustness for performance anomaly detection. In addition, the framework can accurately localize the first root causes, with an average accuracy higher than 0.7 for locating the first four root cause metrics.

Digital Twin (DT) is a prominent technology to utilise and deploy within the healthcare sector. Yet, the main challenges facing such applications are: strict health data-sharing policies, high-performance network requirements, and possible infrastructure resource limitations. In this paper, we address all the challenges by provisioning adaptive Virtual Network Functions (VNFs) to enforce security policies associated with different data-sharing scenarios. We define a Cloud-Native Network orchestrator on top of a multi-node cluster mesh infrastructure for flexible and dynamic container scheduling. The proposed framework considers the intended data-sharing use case, the policies associated, and infrastructure configurations, then provisions Service Function Chaining (SFC) and provides routing configurations accordingly with little to no human intervention. As a result, we provide an adaptive network orchestration for digital health twin use cases, that is policy-aware, requirements-aware, and resource-aware.

This paper examines the Flexible Algorithm (FlexAlgo) for its potential to enable user-driven path control in intra-domain Segment Routing (SR) enabled networks. FlexAlgo is a relatively new approach to intra-domain routing that allows multiple custom algorithms to coexist within a single domain. This capability has the potential to provide users with greater control over the paths their data takes through a network. The research includes a thorough investigation of the FlexAlgo approach, including an examination of its underlying techniques, as well as a practical implementation of a FlexAlgo-based solution. We depict performed experiments where we implemented FlexAlgo in three different scenarios. We also present how we developed an automated tool for users to control traffic steering using preferred metrics and constraints. The results of this investigation demonstrate the capabilities of FlexAlgo as a means of enabling user-driven path control and therefore increase security and trust of users towards the network.

The fifth generation (5G) of wireless networks is set out to meet the stringent requirements of vehicular use cases. Edge computing resources can aid in this direction by moving processing closer to end-users, reducing latency. However, given the stochastic nature of traffic loads and availability of physical resources, appropriate auto-scaling mechanisms need to be employed to support cost-efficient and performant services. To this end, we employ Deep Reinforcement Learning (DRL) for vertical scaling in Edge computing to support vehicular-to-network communications. We address the problem using Deep Deterministic Policy Gradient (DDPG). As DDPG is a model-free off-policy algorithm for learning continuous actions, we introduce a discretization approach to support discrete scaling actions. Thus we address scalability problems inherent to high-dimensional discrete action spaces. Employing a real-world vehicular trace data set, we show that DDPG outperforms existing solutions, reducing (at minimum) the average number of active CPUs by 23% while increasing the long-term reward by 24%.

Cloud computing attracts increasing attention in processing dynamic computing tasks and automating the software development and operation pipeline. In many cases, the computing tasks have strict deadlines. The cloud resource manager (e.g., orchestrator) effectively manages the resources and provides tasks Quality of Service (QoS). Cloud task scheduling is tricky due to the dynamic nature of task workload and resource availability. Reinforcement Learning (RL) has attracted lots of research attention in scheduling. However, those RL-based approaches suffer from low scheduling performance robustness when the task workload and resource availability change, particularly when handling time-critical tasks. This paper focuses on both challenges of robustness and deadline guarantee among such RL, specifically Deep RL (DRL)-based scheduling approaches. We quantify the robustness measurements as the retraining time and investigate how to improve both robustness and deadline guarantee of DRL-based scheduling. We propose MLR-TC-DRLS, a practical, robust Meta Deep Reinforcement Learning-based scheduling solution to provide time-critical tasks deadline guarantee and fast adaptation under highly dynamic situations. We comprehensively evaluate MLR-TC-DRLS performance against RL-based and RL advanced variants-based scheduling approaches using real-world and synthetic data. The evaluations validate that our proposed approach improves the scheduling performance robustness of typical DRL variants scheduling approaches with 97%–98.5% deadline guarantees and 200%–500% faster adaptation.

On the road towards personalised medicine, one of the main challenges is to enforce security and network low-level policies to secure data-sharing. The proposed dynamic framework defines the topology of the service chains to enforce network and security policies by instantiating Virtual Network Functions (VNF's) on the fly via light-weight and easily-deployable containers. In this paper, we profile the resource utilisation of chained VNF's deployed to enable data movement within different healthcare use cases. We provide example configurations that map to a couple of use cases (e-Health record query and heath data streaming), then we monitor and collect CPU utilisation of the different VNF compositions. In the considered policies we can: discard flow, protect (encrypt) and transmit, or allow with no protection. To enforce each policy, we deploy a firewall function (relatively heavy-weight function), encryption function, and a decryption function (light-weight stream cipher). As a result, we analyse the behaviour of the VNF services with various setups, and then we aim to further use this analysis to build the placement heuristic according to available and trusted clusters resources. Subsequently, we will recommend heuristic-based placement based on collected profiling data of the resource usage and limits for high availability, optimal performance, and minimal resource waste.

Utilising programmable infrastructures is a promising approach to support secure data-sharing across healthcare domains of different capabilities in terms of network and security. The EPI1(Enabling Personalised Interventions) framework automates the setup of the underlying infrastructure while considering different requirements communicated by the EPI components, such as logic area generator and policy management system.In our approach to dynamically provide collaborative environments, we containerise network functions (VFs), that are shipped out and instantiated at the edge of the network. We use container-based Virtual Network Function (VNFs) to accomplish fast deployment, high reusability, and low-performance overhead.Traffic interception and redirection through the chain of containerised network functions is a core feature of our framework. In this paper, we focus on the implementation and design of this tool for packet interception and redirection. We evaluate the performance of two approaches: an NGINX-based reverse proxy method and a SOCKS protocol-based method. We benchmark them to determine the overhead compared to a direct data-sharing session with no proxy. We conclude that the reverse proxy performs better in terms of overhead. Nonetheless, the SOCKS-based method works on a lower network level support all traffic types and offer a higher processing rate. We compare the methods according to other performance parameters. Subsequently, the choice of method will depend on the application performance requirements.

A Digital Data Marketplace (DDM) facilitates secure and trustworthy data sharing among multiple parties. For instance, training a machine learning (ML) model using data from multiple parties normally contributes to higher prediction accuracy. It is crucial to enforce the data usage policies during the execution stage. In this paper, we propose a methodology to distinguish programs running inside containers by monitoring system calls sequence externally. To support container portability and the necessity of retraining ML models, we also investigate the stability of the proposed methodology in 7 typical containerized ML applications over different execution platform OSs and training data sets. The results show our proposed methodology can distinguish between applications over various configurations with an average classification accuracy of 93.85%, therefore it can be integrated as an enforcement component in DDM infrastructures.

Recently, Digital Data Marketplaces (DDMs) are gaining wide attention as a sharing platform among different organizations. That is due to the fact that sharing the information and participating in research collaborations play an important role in addressing multiple scientific challenges. To increase trust among participating organizations multiple contracts and agreements should be established in order to determine regulations and policies about who has access to what. Describing these agreements in a general model to be applicable in different DDMs is of utmost importance. In this paper, we present a semantic model for describing the access policies by means of semantic web technologies. In particular, we use and extend the Open Digital Rights Language (ODRL) to describe the pre-established agreements in a DDM.

Nowadays, the cyber, social and physical worlds are increasingly integrating and merging. Especially, combining the strengths of humans and machines helps tackle increasing hard tasks that neither can be done alone. Following this trend, this paper designs a Quality aware Truthful Incentive mechanism for cyber–physical enabled Geographic crowdsensing called Geo-QTI. Different from existing work, Geo-QTI appropriately accommodates the utilities of various stakeholders: requesters, participants and the crowdsourcing platform, and explicitly takes the requesters’ quality requirements, and participants’ quality provision into account. Geo-QTI explicitly includes four components: requester selection, participant selection, pricing and allocation. Requester selection with feasible analysis removes the requesters whose job cannot be completed by all participants or suffers from the monopoly participant (without the participant’s contribution, others cannot cover requesters’ requirement), obtains winning requesters set and determines actual payments. In participant selection phase, the platform aggregates the requested tasks (submitted by all winning requesters) in the sensed geographic area, and chooses the appropriate participants satisfying the winning requesters’ quality requirements with total cost as low as possible. Pricing phase determines the payments to winning participants. The phase of allocation assigns the specific participants to minimally cover the quality requirements of those winning requesters. Rigid theoretical analysis demonstrates Geo-QTI can achieve both requesters’ and participants’ individual rationality and truthfulness, computational efficiency and budget balance for the platform. Furthermore, the extensive simulations confirm our theoretical analysis, and illustrate that Geo-QTI can reduce requesters’ expenses greatly and ensure the fairness of allocation.

To address increasing problems caused by cyber attacks, we leverage Software Defined networks and Network Function Virtualisation governed by a SARNET-agent to enable autonomous response and attack mitigation. A Secure Autonomous Response Network (SARNET) uses a control loop to constantly assess the security state of the network by means of observables. Using a prototype we introduce the metrics impact and effectiveness and show how they can be used to compare and evaluate countermeasures. These metrics become building blocks for self learning SARNET which exhibit true autonomous response.

Cloud Computing has several provisioning models, namely Infrastructure as a service (IaaS), Platform as a service (PaaS), and Software as a service (SaaS). However, cloud users (tenants) have limited or no control over the underlying network resources and services. Network as a Service (NaaS) is emerging as a novel model to bridge this gap. However, NaaS requires an approach capable of modeling the underlying network resources and capabilities in abstracted and vendor-independent form. In this paper we elaborate on SemNaaS, a Semantic Web based approach for supporting network management in NaaS systems. Our contribution is three-fold. First, we adopt and improve the Network Markup Language (NML) ontology for describing NaaS infrastructures. Second, based on that ontology, we develop a network modeling system that is integrated with the existing OpenNaaS framework. Third, we demonstrate the benefits that Semantic Web adds to the Network as a Service paradigm by applying SemNaaS operations to a specific NaaS use case.

Linux container virtualisation is gaining momentum as lightweight technology to support cloud and distributed computing. Applications relying on container architectures might at times rely on inter-container communication, and container networking solutions are emerging to address this need. Containers can be networked together as part of an overlay network, or with actual links from the container to the network via kernel modules. Most overlay solutions are not quite production ready yet; on the other hand kernel modules that can link a container to the network are much more mature. We benchmarked three kernel modules: veth, macvlan and ipvlan, to quantify their respective raw TCP and UDP performance and scalability. Our results show that the macvlan kernel module outperforms all other solutions in raw performance. All kernel modules seem to provide sufficient scalability to be deployed effectively in multi-containers environments.

Environmental research infrastructures (RIs) support data-intensive research by integrating large-scale sensor/observer networks with dedicated data curation services and analytical tools. However the diversity of scientific disciplines coupled with the lack of an accepted methodology for constructing new RIs inevitably leads to incompatibilities between the data models, metadata standards and service descriptions used by different RIs, inhibiting their usefulness for interdisciplinary research. In the absence of a common global ontology of science and infrastructure, these inconsistencies may best be counteracted by selectively bridging the semantics of the various vocabularies, standards and models used by the RIs at present. Open Information Linking for Environmental RIs (OIL-E) was developed within the FP7 project ENVRI to provide a framework for semantic linking of knowledge resources used by different environmental RIs. Built around a multi-viewpoint reference model ENVRI-RM, OIL-E is intended to act as a central exchange for linking information fragments and identifying gaps in the conceptual models of RIs.

Environmental research infrastructures (RIs) support their respective research communities by integrating large-scale sensor/observation networks with data curation services, analytical tools and common operational policies. These RIs are developed as pillars of intra-and interdisciplinary research, however comprehension of the complex, pathologically interconnected aspects of the Earth's ecosystem increasingly requires that researchers conduct their experiments across infrastructure boundaries. Consequently, almost all data-related activities within these infrastructures, from data capture to data usage, needs to be designed to be broadly interoperable in order to enable real interdisciplinary innovation. The Data for Science theme in the EU Horizon 2020 project ENVRIPLUSintends to address this interoperability challenge as it relates to the design, implementation and operation of environmental science RIs, the theme focuses on key issues of data identification and citation, curation, cataloguing, processing, optimization, and provenance, supported by a generic cross-infrastructure reference model.

Storing data in the cloud is becoming a common trend, for both end-customers and data center operators. We propose a method for deciding where to host data storage tasks under the constraint of minimal greenhouse gas emission. The decision on whether to store data locally or store it remotely at a cleaner data center relies on the models for the local and remote data centers and the network connecting them. We conclude that the transport network that connects a local node and a "cleaner" remote data center plays a significant role in the decision of where to store data, and that the frequency of access of the data is an important and related factor.

The Network Service Interface (NSI) has been created as a result of collaborative development of network and application engineers primarily associated with the Research and Education (R&E) community. The NSI allows workflow systems not only to check available service points for a workflow engine to schedule executions, but also to reserve and provide network connections among those service points. The Open Flow technology provides programmability on the network Flow and allows software to define dynamically behaviour of the network. These new features offer data intensive applications new opportunities to optimize the mapping between data Flow patterns and the infrastructure yielding better system level quality. However, they also require the computing support systems effectively capture not only the characteristics of the application workflow but also the controllability of the underlying network. In this paper we discussed the extension of our previous system called Network QoS Planner (NEWQoSPlanner) and investigated how reservation based connection services can be enhanced by dynamic network Flow control. We also discusse how NEWQoSPlanner invokes network services to achieve connection reservation and provisioning, and includes Open Flow to realize dynamic Flow optimization for data intensive workflows.

SUMMARY Ontology development involves people with different background knowledge and expertise. It is an elaborate process, where sophisticated tools for experienced knowledge engineers are available. However, domain experts need simple tools that they can use to focus on ontology instantiation. In this paper, we propose a methodology with a separation of concern between domain experts and knowledge engineers. This separation allows domain experts to focus on information processing and ontology instantiation while providing immediate feedback to the knowledge engineers on usability of the ontology being developed. We have designed and implementedOINTED, an adaptive online ontology instance editor that supports this methodology. We present usage examples ofOINTEDthat highlight three main features: the intuitive visualization of concepts, instances, and relationships within an ontology; the seamless integration in pre‐existing problem solving environment; and the assistance in ontology evolution.OINTEDcomplements existing tools suited for knowledge engineers by enabling immediate feedback and a shorter ontology development life cycle. Copyright © 2012 John Wiley & Sons, Ltd.

This paper discusses the challenges that are imposed by Big Data Science on the modern and future Scientific Data Infrastructure (SDI). The paper refers to different scientific communities to define requirements on data management, access control and security. The paper introduces the Scientific Data Lifecycle Management (SDLM) model that includes all the major stages and reflects specifics in data management in modern e-Science. The paper proposes the SDI generic architecture model that provides a basis for building interoperable data or project centric SDI using modern technologies and best practices. The paper explains how the proposed models SDLM and SDI can be naturally implemented using modern cloud based infrastructure services provisioning model.

This paper describes the Infrastructure and Network Description Language (INDL). The aim of INDL is to provide technology independent descriptions of computing infrastructures. These descriptions include the physical resources and the network infrastructure that connects these resources. The description language also provides the necessary vocabulary to describe virtualization of resources and the services offered by these resources. Furthermore, the language can be easily extended to describe federation of different existing computing infrastructures, specific types of (optical) equipment and also behavioral aspects of resources, for example, their energy consumption. Before we introduce INDL we first discuss a number of modeling efforts that have lead to the development of INDL, namely the Network Description Language, the Network Markup Language and the CineGrid Description Language. We also show current applications of INDL in two EU-FP7 projects: NOVI and GEYSERS. We demonstrate the flexibility and extensibility of INDL to cater the specific needs of these two projects.

The Green Clouds project in the Netherlands investigates a system-level approach towards greening High-Performance Computing (HPC) infrastructures and clouds. In this paper we present our initial results in profiling virtual machines with respect to three power metrics, i.e. power, power efficiency and energy, under different high performance computing workloads. We built a linear power model that represents the behavior of a single work node and includes the contribution from individual components, i.e. CPU, memory and HDD, to the total power consumption of a single work node. Our results could be part of a power characterization module integrated into clusters' monitoring systems, future Green Clouds energy-savvy scheduler would use this monitoring system to support system-level optimization.

This paper presents our work toward organizing and managing various forms of federations of virtualized infrastructures. We adopt the Ponder2 policy framework and the SMC architecture as a powerful engineering approach, which we apply to semantic-aware management of federations of Future Internet (FI) virtualized infrastructures. To cater for context-awareness, we plan for a common information model, based on the Network Description Language (NDL), capturing a common set of abstractions of virtualized resources and services, nodes, routers and switches, custom network topologies with specific bandwidth demands, etc. To handle management of generic complex federated environments, we employ structural patterns to model federations as graphs, whose vertices represent SMCs and edges denote the type of relationship between them. We give an illustration of such structures corresponding to existing FI experimental platforms in the US and Europe and we provide examples containing inter-domain management responsibilities as missions. Finally, we propose to augment the Ponder2 framework with single & multi-domain resource provisioning capabilities, enabling efficient sharing of virtualized networked facilities among federation users.

Semantic web technologies provide a standardised mechanism for describing and accessing the services of underlying infrastructure. These technologies facilitate the inclusion of the quality of network services in the control loop of high level applications and allow applications to tune the system level performance with additional quality dimensions. However, the descriptions of a large infrastructure are often composed and maintained by different parties and can have different levels of details because of the administration policies. These facts make the development of high level applications unnecessarily difficult. We present a preprocessing framework to hide these difficulties from high level application developers by transforming, integrating, and filtering raw descriptions of the infrastructure into proper information content that these applications need.

The amount of computing resources currently available on Clouds is large and easily available with pay per use cost model. E-Science applications that need on-demand execution benefit from Clouds, because no permanent computing resources to support peak demand has to be acquired. In this paper, we present AMOS, a system that automates creation and management of temporary Grids on a Cloud to execute (parts of) application workflows. We performed experiments with AMOS and a representative e-Science application on a research Grid and on the Amazon EC2 Cloud. The results show that AMOS is a viable approach to manage and execute e-Science applications in a flexible Grid environment and to explore novel mechanisms that allow optimal utilization of Cloud resources. Furthermore, we consider AMOS as a step towards an operating system for (virtual) infrastructures that enables Grid applications to control their computational resources at run-time.

The introduction of optical private networks (lightpaths) has significantly improved the capacity of long distance network links, making it feasible to run large parallel applications in a distributed fashion on multiple sites of a computational grid. Besides offering bandwidths of 10 Gbit/s or more, lightpaths also allow network connections to be dynamically reconfigured. This paper describes our experiences with running data-intensive applications on a grid that offers a (manually) reconfigurable optical wide-area network. We show that the flexibility offered by such a network is useful for applications and that it is often possible to estimate the necessary network configuration in advance.

Current research networks allow end users to build their own application-specific connections (lightpaths) and optical private networks (OPNs). This requires a clear communication between the requesting application and the network. The network description language (NDL) is a vocabulary designed to describe optical networks based on the resource description framework (RDF). These descriptions aid applications in querying the capabilities of the network and allow them to clearly express requests to the network. This article introduces NDL and shows its current applications in optical research networks.

Several research networks around the world are implementing hybrid networks, that provide end-users with traditional routed IP together with lightpaths. These paths are dynamically configured at user request and network provisioning systems must have topology information, both intra- and inter-domain.We developed the Network Description Language (NDL), based on RDF, a semantic web technique. This language can be used to describe hybrid networks, so that different administrative domains can share and correlate topology information. It supports the end-user to express a lightpath reservation request, and helps the service provider to validate the feasibility of requests. It facilitates generation and exchange of network maps by allowing automatic correlation of information across domains.Our first application ground is GLIF, a collaboration promoting co-operation for Lambda Networking. Several tools for automatic provisioning are in development. However, these tools lack a common network description, which NDL can provide.