Real-time processing of inference tasks generated by resource-constrained devices in an Edge Computing environment demands carefully designed solutions that guarantee the performance and achieve a high level of accuracy. To reduce transmission time, inference tasks are often compressed to reduce the transmission time and offloaded to the edge infrastructure for parallel batch processing in GPUs. In this setting, an interesting tradeoff arises, characteristic of Approximate Computing, where the quality of inference and the system’s end-to-end latency are competing objectives. In this paper, we formulate a joint optimization problem to maximize the quality of inference while minimizing the overall latency for the GPU-enabled batch processing of inference applications. The optimization problem is NP-hard, and we split it into two subproblems to obtain optimal values for the compression of offloaded tasks and select the offloading strategy that minimizes the total latency. By carefully examining the results of the compression problem, we identify that compressing the tasks in such a way to arrive simultaneously for remote processing significantly increases the performance of batch processing. To compute an offloading strategy, we employ a semidefinite relaxation (SDR)-based approach and a randomized mapping to obtain feasible solutions. Therefore, we design an iterative alternating algorithm to solve both problems and obtain a near-optimal solution in polynomial complexity. Simulation results indicate that the proposed framework outperforms all compared solutions by reducing the total cost by 50%.

The dynamic workload demands of smart city applications hosted on edge infrastructures require the development of advanced scaling mechanisms. Recent studies proposed single-application autoscaling solutions based on various technical approaches. However, for edge infrastructures with limited resource availability, it is essential to simultaneously manage heterogeneous application requirements, aiming at optimal resource allocation and minimal operational costs. This study introduces a multi-application hierarchical autoscaling framework for Kubernetes Edge Clusters. An application-based mechanism nominates the best applications’ deployments based on workload prediction and several criteria that guarantee the application’s performance while minimizing the infrastructure provider’s cost. For the joint application orchestration, an aggregation mechanism composes the candidate scaling solutions for the cluster. Then, a cluster autoscaling mechanism, based on the Analytic Hierarchy Process, undertakes the cluster’s scaling decision to optimize the resource allocation and energy consumption of the cluster. The evaluation illustrates the benefits of the proposed scaling strategy, achieving significant improvement in the average allocated resources and energy consumption compared to single-application approaches.

The evolution of the Industrial Internet of Things (IIoT) and edge computing enables resource-constrained mobile robots to offload the computationally intensive localization algorithms. Naturally, utilizing the remote resources of an edge server to offload these tasks encounters the challenge of a joint co-design in communication, control, estimation, and computing infrastructure. We introduce a set-based estimation offloading framework, for the specific case of the navigation of a unicycle robot toward a target position. The robot is subject to modeling and measurement uncertainties, and the estimation set is calculated using overapproximation techniques that alleviate additional computations. A switching set-based control mechanism provides accurate navigation and triggers more precise estimation algorithms when needed. To guarantee the convergence of the system and optimize the utilization of remote resources, a utility-based offloading mechanism is designed, which takes into account both the dynamic network conditions and the available computing resources at the network edge. The performance of the proposed framework is demonstrated through simulations and comparison with alternative offloading schemes.

Maximizing the performance of modern applications requires timely resource management of the virtualized resources. However, proactively deploying resources for meeting specific application requirements subject to a dynamic workload profile of incoming requests is extremely challenging. To this end, the fundamental problems of task scheduling and resource autoscaling must be jointly addressed. This paper presents a scalable architecture compatible with the decentralized nature of Kubernetes [1], to solve both. Exploiting the stability guarantees of a novel AIMD-like task scheduling solution, we dynamically redirect the incoming requests towards the containerized application. To cope with dynamic workloads, a prediction mechanism allows us to estimate the number of incoming requests. Additionally, a Machine Learning-based (ML) Application Profiling Modeling is introduced to address the scaling, by co-designing the theoretically-computed service rates obtained from the AIMD algorithm with the current performance metrics. The proposed solution is compared with the state-of-the-art autoscaling techniques under a realistic dataset in a small edge infrastructure and the trade-off between resource utilization and QoS violations are analyzed. Our solution provides better resource utilization by reducing CPU cores by 8% with only an acceptable increase in QoS violations.

Autoscaling resources in a power-efficient way is essential to enable Green Computing resource management solutions. The development of dynamic resource provisioning techniques could lead to the minimization of power consumption and simultaneously guarantee high quality of service (QoS) inline with the workload demand. In this work, we introduce AHP4HPA, an autoscaling framework for Kubernetes Clusters, which is aligned with the Kubernetes architecture and state-of-the-art practices. We define resource profiles, namely a mapping between the QoS and the computing resources, to maximize the performance. Furthermore, Analytic Hierarchy Process (AHP) is exploited to dictate the scaling decision of the resources under various Key Performance Indicators (KPIs) toward power optimization of the allocated resources. To guarantee maximum performance of the deployed image classification application, an ARIMA model is dedicated to providing predictions regarding the incoming workload traffic. The framework is evaluated against a realistic dataset in a small-scale testbed. Numerical results indicate at least a 9% reduction of the average energy consumption when compared to other state of the art techniques.

Network slicing has been at the forefront of 5G network research, with various slicing orchestration architectures seeking to reap the benefits of slicing for the enhanced performance and reliability of 5G (and beyond) network services. In this context, cross-slice communication (CSC) has drawn significant attention, since CSC can foster interactions among services deployed in co-located slices, lowering the barrier for the consumption of services.To capitalize the benefits of CSC (e.g., reduced latency and cost), CSC should be established with the highest degree of co-location and also in a secure and policy-compliant manner. To this end, we present an orchestration framework that fulfills all main technical requirements for CSC instantiation. In this respect, we elaborate on the CSC instantiation workflows and shed light into the cross-layer interactions that span our proposed CSC orchestrator, the Network Function Virtualization Orchestrator (NFVO) and the Virtualized Infrastructure Manager (VIM). Our experimental results indicate that our proposed CSC orchestration framework introduces a negligible performance overhead and also incurs a minimal latency inflation compared to a direct form of inter-slice communication without any provision for security and resource isolation.

The emergence of Industrial Internet of Things (IIoT) requires the interconnection between robots, sensors, and the underlying network and computing infrastructure. Edge Robotics has emerged as a flexible paradigm that enables resource-constrained mobile robots to offload computationally intensive tasks of time/mission-critical applications. In this context, Edge Computing is essential for providing additional resources towards confronting the stringent performance specifications. This article presents the architectural concepts and capabilities of the NETMODE testbed, member of the Fed4FIRE+ federation, for the state-of-the-art experimentation with robotic applications. An evaluation of the proposed architecture is conducted using a SLAM algorithm which is a compute-intensive application.

In the era of 5G networks, the demand for high quality service provisioning is growing extremely fast. The enabling of Network Function Virtualization and Network Slicing in the scope of 5G network aims to meet the strict requirements of various business cases. Alongside, the complexity of deployment such services becomes also higher, regarding the differences between infrastructure capabilities and the plethora of various individual requirements. This fact makes the selection of the appropriate infrastructure for slice deployment a complex, but also, a major process, as the optimization of the selection leads to the satisfaction of the user and the better resource allocation from the provider's perspective. In this work, an Edge PoP Selection framework for network slice deployment is proposed. This framework takes into account the user's hard and soft requirements and performs a two-stage selection. The selection of the appropriate infrastructure is based on a multi-criteria decision making method. The proposed framework is evaluated and compared with simple filtering and single-objective selection approaches. The promising results show the importance of the two stage framework in order to simultaneously meet the user's requirements and the optimal utilization of the resources.

The increasing resource demand and diversity of network services is taken under serious consideration by the various stakeholders, driving the architecture design of 5G (and beyond) networks. Network slicing, as a prominent aspect of next-generation network architectures, aims at satisfying the diverse service requirements in terms of throughput, latency, reliability, and/or security. However, the prevailing way of slice provisioning, i.e., in the form of isolated bundles of computing, storage, and network resources, makes cross-slice communication inefficient, especially at the network edge. This inevitably hinders opportunities for Business-to-Business (B2B) synergies at the event of service co-location. In this paper, we study this novel aspect of network slicing, i.e., cross-slice communication (CSC). We particularly promote a form of optimized CSC, at which two co-located slices can establish peering in a secure and controlled manner, by confining peering traffic within the boundaries of the datacenter, while still preserving the important aspect of resource isolation. Such optimized CSC can foster synergies between service providers without additional latency or traffic in the backhaul/transport network. In this context, we investigate various ways to establish optimized CSC at edge computing infrastructures, based on functionalities offered by state-of-the-art management and orchestration (MANO) frameworks, such as OpenSourceMANO.

The evolution of Internet of Things (IoT) has sparked significant research interest in edge computing. Within this scope and given the ever-increasing number of IoT and mobile devices, computation offloading is emerging as a cutting-edge and significant research area with enormous potential and practical applications.In this respect, we present the architecture design and experimental evaluation of an orchestration framework for smart computation offloading from IoT or mobile devices to edge cloud servers. The proposed orchestration platform, namely COSMOS, includes control-plane components for workload prediction, load balancing, and admission control. COSMOS is particularly tailored to the needs of an object identification service that receives images from a multitude of Points of Interest (PoIs), performs object identification using a trained model (based on Tensorflow), calculates the prediction accuracy, and finally returns to the end-users the identification outcome and accuracy along with useful information about the identified object. COSMOS has been deployed and evaluated in a large-scale experimental facility that employs OpenStack and OpenSourceMANO (OSM) for Network Function Virtualization (NFV) orchestration. Our experimental results indicate the feasibility of computation offloading for this object identification service and further uncover useful insights in terms of performance and scalability.

Although mobile devices today have powerful hardware and networking capabilities, they fall short when it comes to executing compute-intensive applications. Computation offloading (i.e., delegating resource-consuming tasks to servers located at the edge of the network) contributes toward moving to a mobile cloud computing paradigm. In this work, a two-level resource allocation and admission control mechanism for a cluster of edge servers offers an alternative choice to mobile users for executing their tasks. At the lower level, the behavior of edge servers is modeled by a set of linear systems, and linear controllers are designed to meet the system’s constraints and quality of service metrics, whereas at the upper level, an optimizer tackles the problems of load balancing and application placement toward the maximization of the number the offloaded requests. The evaluation illustrates the effectiveness of the proposed offloading mechanism regarding the performance indicators, such as application average response time, and the optimal utilization of the computational resources of edge servers.

Unmanned Aerial Vehicles (UAV) facilitate the development of Internet of Things (IoT) ecosystems for smart city and smart environment applications. This paper proposes the adoption of Edge and Fog computing principles to the UAV based forest fire detection application domain through a hierarchical architecture. This three-layer ecosystem combines the powerful resources of cloud computing, the rich resources of fog computing and the sensing capabilities of the UAVs. These layers efficiently cooperate to address the key challenges imposed by the early forest fire detection use case. Initial experimental evaluations measuring crucial performance metrics indicate that critical resources, such as CPU/RAM, battery life and network resources, can be efficiently managed and dynamically allocated by the proposed approach.

Managing the performance of virtualized web applications is a critical problem for service providers. Meeting their Service Level Objectives (SLOs), such as response time, in a dynamic environment(dense load, variable capacity), is a challenging task. In this paper, we propose an autonomous scheme for admission control in cloud services aiming at preventing overloading, guaranteeing target response time and dynamically adapting the admitted workload to compensate for changes in system capacity. We employ an adaptive feedback control scheme alongside with a queue model of the application. We show that our solution effectively delimits the response time of requests maintaining high throughput levels while it adapts to variations of system capacity.