Fifth Generation (5G) networks and beyond are envisioned to provide user-focused communications, supporting diverse services with enhanced Quality of Service (QoS). Pivotal to this evolution is user equipment, which is increasingly performing advanced computational tasks beyond the edge of the network, known as the Extreme-Edge. Seamless integration of Extreme-Edge devices (EEDs) into the 5 G framework is however hindered, due to challenges in terms of device management, resource restrictions and interoperability issues. To address these barriers, we realize the Extreme-Edge Orchestrator (EEO), a management and orchestration framework enabling the extension of the 5 G cloud-to-edge continuum towards the Extreme-Edge. The EEO enables real-time resource monitoring and lifecycle management of network applications, including Artificial Intelligence/Machine Learning (AI/ML) tasks, deployed on EEDs. Unlike existing theoretical studies, our solution is deployed on an operational research-center-wide 5 G testbed and evaluated using an AI/ML-based network QoS prediction application, in the automotive domain. Our results show that the EEO supports efficient resource utilization, dynamic EED selection under device mobility scenarios and maintains robust service performance under computational stress-demonstrating its capability to support next-generation network services.

The advent of 5G and beyond systems is expected to shape the automotive vertical, as safety-critical vehicular applications rely on the network to meet their stringent Quality of Service (QoS) requirements. Predictive QoS (pQoS) has been proposed as a mechanism that allows automotive applications to proactively adapt in view of forthcoming QoS changes. Although pQoS is typically facilitated via classical (centralized) Machine Learning (ML) methods, the demand for data privacy has led to the emergence of distributed ML schemes. Efficient training of ML models however requires large volumes of (kinematic-state and connectivity) QoS data, so as to capture the involved spatio-temporal effects.To that end we hereby present and publicly share NordicDat, a QoS dataset collected during a two-week measurement campaign, driving across three European countries. NordicDat contains over 90K samples of physical layer, network and mobility-related features. Contrary to prior works, it includes multiple instances of cross-boarder roaming, diverse vehicle speed profiles and radio access technologies (generations). Further, we provide a thorough NordicDat data analysis, highlighting the dependencies between the NordicDat’s features and the resulting QoS values (throughput, delay). To showcase its broad usability, we train pQoS ML models over NordicDat in classical and distributed fashion. Our results demonstrate for the first time the viability of distributed pQoS with real-word data, which achieves similar (within a margin of 10%) accuracy to that of classical ML, cropping privacy-preserving benefits.

Edge computing is a recent paradigm where the processing takes place close to the data sources. It therefore reduces latency and saves bandwidth compared to traditional cloud computing. The latter can continue to play a supportive role. Edge-cloud computing provides benefits in many use cases including distributed computation algorithms, where the processing is divided into a number of tasks that are executed in parallel on different equipment. An important relevant challenge is to allocate the appropriate resources to process the data that are continuously generated from user devices. The issue becomes more complicated when we take into account the variations in the volume of the generated data as a function of time. In this paper we present a resource allocation algorithm for distributed computation with emphasis on machine learning algorithms. We consider that the resource requirements vary with time in a semi-static way that exhibits some daily pattern. We distinguish between periodic (expected) variations that occur during the day, and sporadic variations due to unexpected events. We propose an Integer Linear Programming algorithm to allocate the periodic resource requirements. To handle the non-periodic requirements, we consider a suitable prediction algorithm coupled with a reconfiguration algorithm that allocates the predicted required resources. Our results indicate that our proposal outperforms traditional allocation algorithms in terms of resource utilization, monetary cost and achieved accuracy.

QoS forecasting for cellular vehicular communications allows cooperative, connected and automated mobility applications to tailor their behavior to the expected communication conditions on the road. In a nutshell, vehicles may, for example, execute cooperative maneuvers if the communication quality of service is only above a certain quantitative level whereas if not they revert to the individual autonomous mode. In this paper, we propose and show empirical methods for estimating packet-based QoS metrics obtained from 5G network measurements with a direct application to vehicular applications. As many distributed vehicular applications possess strict QoS requirements, we focus here on bounding packet-based statistical QoS quantiles, specifically for latency and loss. Our approach is based on training regression neural networks in a federated learning fashion and show that it can obtain predictions on par with centralized training without the vehicles needing to transmit raw measurement data. In contrast to QoS prediction using physical layer information, we briefly discuss the embedding of such much simpler application-level service within the 5G architecture. We also validate our approach through recovering classical closed-form delay quantiles that are obtained from analytical models of simple queueing systems. We show that our approach goes beyond these simple models in that it provides quantile estimates for the complex scenario of cellular vehicle communications and under different application traffic patterns including empirical data traffic traces as well as 5G testbed measurements.

Edge computing has emerged as a computing paradigm where the application and data processing takes place close to the end devices. It decreases the distances over which data transfers are made, offering reduced delay and fast speed of action for general data processing and store/retrieve jobs. The benefits of edge computing can also be reaped for distributed computation algorithms, where the cloud also plays an assistive role. In this context, an important challenge is to allocate the required resources at both edge and cloud to carry out the processing of data that are generated over a continuous (“infinite”) time horizon. This is a complex problem due to the variety of requirements (resource needs, accuracy, delay, etc.) that may be posed by each computation algorithm, as well as the heterogeneous resources’ features (e.g., processing, bandwidth). In this work, we develop a solution for serving weakly coupled general distributed algorithms, with emphasis on machine learning algorithms, at the edge and/or the cloud. We present a dual-objective Integer Linear Programming formulation that optimizes monetary cost and computation accuracy. We also introduce efficient heuristics to perform the resource allocation. We examine various distributed ML allocation scenarios using realistic parameters from actual vendors. We quantify trade-offs related to accuracy, performance and cost of edge/cloud bandwidth and processing resources. Our results indicate that among the many parameters of interest, the processing costs seem to play the most important role for the allocation decisions. Finally, we explore interesting interactions between target accuracy, monetary cost and delay.

Edge computing has emerged as a paradigm for local computing/processing tasks, reducing the distances over which data transfers are made. Thus, an opportunity is presented for data transfer-intensive, distributed machine learning. In this paper we develop a solution for serving distributed Machine Learning (ML) training jobs at the edge– cloud continuum. We model the specific requirements of each ML job, and the features of the edge and cloud resources. Next, we develop an Integer Linear Programming algorithm to perform the resource allocation. We examine different scenarios (different processing and bandwidth costs) and quantify tradeoffs related to performance and cost of edge/cloud bandwidth and processing resources. Our simulations indicate that even though there are many parameters that determine the allocation, the processing costs seem to play on average the most important role. The cloud b/w costs can be significant in certain scenarios. Finally, in certain examined cases, significant monetary benefits can be achieved through the collaboration of both edge and cloud resources when compared to using exclusively edge or cloud resources.

The convergence of Machine Learning (ML) with the edge computing paradigm has paved the way for distributing processing-heavy ML tasks to the network's extremes. As the edge deployment details still remain an open issue, distributed ML schemes tend to be network-agnostic; thus, their effect on the underlying network's resource consumption is largely ignored.In our work, assuming a network tree structure of varying size and edge computing characteristics, we introduce an analytical system model based on credible real-world measurements to capture the end-to-end consumption of ML schemes. In this context, we employ an edge-based (EL) and a federated (FL) ML scheme and in-depth compare their bandwidth needs and energy footprint against a cloud-based (CL) baseline approach. Our numerical evaluation suggests that EL exhibits a minimum of 25% bandwidth-efficiency compared to CL and FL, if employed by a few nodes higher in the edge network, while halving the network's energy costs.

The proliferation of machine learning (ML) applications has lately witnessed a considerable shift to more distributed settings, even reaching hand-held mobile devices; there, contrary to typical Centralized learning (CL) whereby the involved (large amounts of) training data are centrally gathered to train models, the load of training tasks is distributed across a set of capable mobile learners at the expense of their own energy. The idea of Federated learning (FL) has emerged as a privacy-preserving mechanism suggesting that the ML model parameters rather than data, are sent over the network to a central point of aggregation. However, when relaxing the privacy concerns, the debate strongly relates to the available network resources. Interestingly, the sofar theoretical or even experimental comparison of the two approaches overlooks network conditions and remains of low realism. In this work we rely on past measurement studies to introduce a realistic system model that accounts for all involved mobile network conditions such as bandwidth and data availability (af-fecting training accuracy and model aggregation) as well as user mobility patterns (affecting data loss). A dedicated simulation framework we have developed replays rich mobile-traces allowing for a comprehensive comparison of the two ML approaches over a large set of training data shedding light on network-resources utilization, energy efficiency and training convergence. Intuitively, our results suggest that the ratio between the employed raw data and the corresponding ML model shapes the conditions under which FL acts as a network-efficient alternative to CL. Interestingly enough, asymmetry in data availability across users as well as their varying number are shown to hardly affect the FL approach in traffic and energy needs, pointing both to its promising potential and the need for further research.

Mobile edge-cloud (MEC) aims to support low latency mobile services by bringing remote cloud services nearer to mobile users. However, in order to deal with dynamic workloads, MEC is deployed in a large number of fixed-location micro-clouds, leading to resource wastage during stable/low workload periods. Limiting the number of micro-clouds improves resource utilization and saves operational costs, but faces service performance degradations due to insufficient physical capacity during peak time from nearby micro-clouds. To efficiently support services with low latency requirement under varying workload conditions, we adopt the emerging network function virtualization (NFV)-enabled MEC, which offers new flexibility in hosting MEC services in any virtualized network node, e.g., access points, routers, etc. This flexibility overcomes the limitations imposed by fixed-location solutions, providing new freedom in terms of MEC service-hosting locations. In this paper, we address the questions on where and when to allocate resources as well as how many resources to be allocated among NFV-enabled MECs, such that both the low latency requirements of mobile services and MEC cost efficiency are achieved. We propose a dynamic resource allocation framework that consists of a fast heuristic-based incremental allocation mechanism that dynamically performs resource allocation and a reoptimization algorithm that periodically adjusts allocation to maintain a near-optimal MEC operational cost over time. We show through extensive simulations that our flexible framework always manages to allocate sufficient resources in time to guarantee continuous satisfaction of applications' low latency requirements. At the same time, our proposal saves up to 33% of cost in comparison to existing fixed-location MEC solutions.

Building on the adoption of the Network Functions Virtualization (NFV) and Software Defined Networking (SDN) paradigms, 5G networks promise distinctive features including the capability to support multi-tenancy. Virtual Network Operators (VNOs) are expected to co-exist over the shared infrastructure, realizing their network functionality on top of virtualized resources. In this context, we observe the emerging opportunity for establishing synergies between co-located tenants of the infrastructure. In this work, motivated by the rapidly increasing mobile content delivery traffic, we focus on synergies in the form of cache peering relationships between co-located VNOs. Caches co-located within the same (micro-)data centers, benefit from content opportunistically cached at their peers, taking advantage of the shared nature of the infrastructure to reduce latencies and traffic overheads. In this paper, we take a close look at the management and orchestration requirements of the envisioned services, illustrating our on-going approach.

Information-Centric Networking (ICN) has emerged as an interesting approach to overcome many of the limitations of legacy IP-based networks. However, the drastic changes to legacy infrastructure required to realise an ICN have significantly hindered its adoption by network operators. As a result, alternative deployment strategies are investigated, with Software-Defined Networking (SDN) arising as a solution compatible with legacy infrastructure, thus opening new possibilities for integrating ICN concepts in operators' networks. This paper discusses the seamless integration of these two architectural paradigms and suggests a scalable and dynamic network topology bootstrapping and management framework to deploy and operate ICN topologies over SDN-enabled operator networks. We describe the designed protocol and supporting mechanisms, as well as the minimum required implementation to realize this inter-operability. A proof-of-concept prototype has been implemented to validate the feasibility of the approach. Results show that topology bootstrapping time is not significantly affected by the topology size, substantially facilitating the intelligent management of an ICN-enabled network.

Name resolution is at the heart of Information-Centric Networking (ICN), where names are used to both identify information and/or services, and to guide routing and forwarding inside the network. The ICN focus on information, rather than hosts, raises significant concerns regarding the scalability of the required Name Resolution System (NRS), especially when considering global scale, inter-domain deployments. In the route-by-name approach to NRS construction, name resolution and the corresponding state follow the routing infrastructure of the underlying inter-domain network. The scalability of the resulting NRS is therefore strongly related to the topological and routing characteristics of the network. However, past work has largely neglected this aspect. In this paper, we present a detailed investigation and comparison of the scalability properties of two route-by-name inter-domain NRS designs, namely, DONA and CURLING. Based on both real, full-scale inter-domain topology traces and synthetic, scaled-down topologies, our work quantifies a series of important scalability-related performance aspects, including the distribution of name-resolution state across the Internet topology and the associated processing and signaling overheads. We show that by avoiding DONA's exchange of state across peering links, CURLING results in deployment costs proportional to the total number of downstream customers of each Autonomous System. This translates to a 62-fold global state size reduction, at the expense of a 2.78-fold increase in lookup processing load, making CURLING a feasible approach to ICN name resolution.

Inspired by the proliferation of content-centric applications in the Internet, Information-Centric Networking (ICN) has emerged as a promising networking paradigm. Focusing on the delivery of content instead of the pairwise communication between end-hosts, ICN inherently supports location-independent content/information distribution, through the means of in-network caching and multicast; as well as mobile computing. However, so far the vast majority of ICN research efforts have mostly focused on the design of sound and scalable architectures and protocols for the current Internet application landscape. In this paper, we revisit ICN in the context of a radically different application environment of smart grids and in particular, the case of smart charging of electric vehicles. Based on a thorough description of the currently forming application environment in the Netherlands, we highlight the inefficiencies resulting from a host-centric model. We then show how ICN can address these limitations and ultimately support quality and security in such application environment. Besides qualitative benefits, our preliminary analysis also demonstrates that ICN can substantially reduce communication and security complexity, thus fostering the development and widespread adoption of the smart charging application.

The proposed shift from host-centric to information-centric networking (ICN) has triggered extensive research in the area of content naming. Efforts have so far focused on the scalability and security properties that can make content objects routable and self-certifying. In this paper, we argue that the information that is exposed through explicitly naming content objects has been overlooked, although several operational and performance issues depend on the information that a name holds. We therefore revisit content naming design decisions taking into account information exposure and deployability of the ICN paradigm.

In the past few years numerous P2P file-sharing and content distribution systems have been designed, implemented, and evaluated via simulations, real world measurements, and mathematical analysis. Yet, only few of them have stood the test of time and gained wide user acceptance. BitTorrent is the one that holds the lion's share among them and the reasons behind its success have been studied to a great extent with interesting results. Nevertheless, even though P2P content distribution remains one of the most active research areas, little progress has been made towards the study of the BitTorrent protocol (and its variations), in a fully controllable and realistic simulation environment. In this paper we describe and analyze a full-featured and extensible implementation of BitTorrent for the OMNeT++ simulation platform. Moreover, since we aim at realistic simulations, we present our enhancements on a popular conversion tool for practical Internet topologies, as well as our churn generator that is based on the analysis of real BitTorrent traces. Finally, we set forth the results from the evaluation of our prototype implementation regarding resource demands under different simulation scenarios.

The use of licensed spectrum for wireless communication is driven by the need to control interference between different operators. However, with this mode of regulation, spectrum utilization is far from efficient and the growth of wireless networks is hindered by the shortage of free frequency bands and the vast investments for the acquisition of a license. In view of this situation, we present an alternative evolution path for the unobstructed growth of wireless networks and the efficient use of spectrum. The proposed architecture is based on the use of unlicensed spectrum and the open access of users to all public networks without prior contracts with operators. We highlight and discuss the inherent technical challenges that must be tackled before the proposed solution can be realized. Special attention is paid to the inherent need for alternative interference mitigation strategies.

The introduction of mobile devices and wireless communications in the context of the Grid computing paradigm has recently drawn the attention of the research community. We point out fundamental issues and problems emerging from this introduction and further describe our ongoing work for the investigation of key design decisions and optimizations.

Recent advances in mobile communications and computing and strong interest of the scientific community in the Grid have led to research into the Mobile Grid. We discuss various approaches proposed in the literature and try to point out the fundamental issues and problems emerging from the introduction of mobile devices and wireless communications in the context of the Grid computing paradigm. We further propose an architecture for the realization of a Mobile Grid and investigate key design decisions and optimizations.

Recent advances in mobile communications and computing and strong interest of the scientific community in the Grid have led to research into the Mobile Grid. We discuss various approaches proposed in the literature and try to point out the fundamental issues and problems emerging from the introduction of mobile devices and wireless communications in the context of the Grid computing paradigm. We further describe an architecture for the realization of a Mobile Grid and investigate key design decisions and optimizations.