Umeå University's logo

As microservice-based systems scale across thecloud-edge continuum, traditional centralized scheduling mecha-nisms increasingly struggle with latency, coordination overhead,and fault tolerance. This paper presents a new architectural di-rection: leveraging service mesh sidecar proxies as decentralized,in-situ schedulers to enable scalable, low-latency coordination inlarge-scale, cloud-native environments. We propose embeddinglightweight, autonomous scheduling logic into each sidecar, allow-ing scheduling decisions to be made locally without centralizedcontrol. This approach leverages the growing maturity of servicemesh infrastructures, which support programmable distributedtraffic management. We describe the design of such an archi-tecture and present initial results demonstrating its scalabilitypotential in terms of response time and latency under varyingrequest rates. Rather than delivering a finalized scheduling algo-rithm, this paper presents a system-level architectural directionand preliminary evidence to support its scalability potential.

Accurate future workload prediction is an essential step for proactive resource allocation and efficient provisioning in cloud computing environments. Deep learning strategies have proven successful for this task, but they face challenges due to the high dimensionality of monitoring data, extensive preprocessing requirements, and computational overhead. In this paper, we propose a hybrid framework that integrates autoencoders for workload compression with Long Short-Term Memory (LSTM) networks for time-series forecasting. Unlike prior studies, our approach systematically analyzes the trade-off between compression ratio and predictive accuracy, demonstrating how dimensionality reduction can improve both scalability and robustness. Thereby reducing the computational burden associated with processing massive-scale monitoring data. Experiments conducted on both synthetic and real-world datasets demonstrate that the proposed method achieves up to 60% data compression with minimal reconstruction loss, while also improving prediction accuracy compared to baseline LSTM models. We evaluate the overall performance of the framework using various metrics, including data reduction ratio, prediction accuracy, and the effects of different compression stages on predictive performance. Additionally, we quantify the computational savings in terms of CPU usage, memory footprint, and training/inference times, confirming the framework's feasibility for real-world deployment. These results underscore the potential of integrating compression and prediction to achieve scalable, accurate, and resource-efficient management of cloud workloads.

The Cloud-Edge Continuum systems are inherently complex and massive, often featuring federated multi-provider stakeholders (e.g. cloud/edge service providers, energy providers), heterogeneous platforms, and dynamic infrastructures; this significantly increases the complexity of developing, deploying, and managing applications. The Serverless computing offers a powerful tool to simplify and speed up the Continuum application development. However, existing scheduling mechanisms for Serverless platforms focus primarily on performance metrics such as latency, model accuracy, and throughput, often neglecting critical factors such as energy efficiency and sustainability. This gap is further exacerbated in Continuum environments, where computational nodes may rely on unpredictable and intermittent green energy sources, leading to availability bottlenecks and energy constraints. This work investigates the design of a decentralized green energy-aware approach for scheduling Serverless functions across the Cloud-Edge Continuum. To achieve this, we introduce a formal model of the green energy-aware workload scheduling problem. We then develop a consensus-based upper confidence bound (UCB) approach for cooperative multi-agent reinforcement learning (MARL) that leverages distributed agents to consider energy awareness and quality-of-service (QoS) requirements of different functions into their scheduling decisions. To demonstrate the practicality of our approach, we implement a real-world prototype using a cluster of Raspberry Pis, Cloud servers, Kubernetes, and OpenFaaS. Experimental results show that our approach maximizes the green energy utilization by (44%) and reduces total latency by (25%) compared to the centralized technique, highlighting its energy efficiency, scalability, and overall sustainability in Continuum settings.

With the exponential growth of edge devices, the cloud edge continuum provides a natural evolution to the centralised cloud architecture to overcome the bottlenecks created by the growing data that devices generate. Resource constrained edge devices need to be able to offload computational tasks to the cloud edge continuum. Providing resources located at the edge, close to resource-constrained devices, allows devices to offload on demand potentially complex functions with low latency and response time requirements. The COGNIT framework introduces the novel concept of function as a services (FaaS) at the edge and novel AI techniques for cloud-edge management. In this paper, we show how the novel edge FaaS model can be used to offload critical security functions from edge devices and enable them to protect themselves even though they don't have the resources for such protection. The paper's main contribution is to show how the edge FaaS model enables to design multi-layer protection models between edge devices and the cloud-edge continuum AI-based orchestrator. In this model the edge device provides a first layer of defense using application knowledge to protect itself whereas the AI-based orchestrator provides a second layer of defense that is more generic because it does not know much about the edge application. The layered protection model is illustrated and validated on a cybersecurity case study where AI-based anomaly detection is deployed at the edge to secure mobile devices and detect anomalies as early and quickly as possible. This second contribution of the paper shows how continuous security anomaly detection can be designed as multiple functions that are triggered by monitored events to provide continuous detection at the edge for all events.

The increasing complexity of distributed computing environments necessitates efficient resource management strategies to optimize performance and minimize resource consumption. Although proactive horizontal autoscaling dynamically adjusts computational resources based on workload predictions, existing approaches primarily focus on improving workload resource consumption, often neglecting the overhead introduced by the autoscaling system itself. This could have dire ramifications on resource efficiency, since many prior solutions rely on multiple forecasting models per compute node or group of pods, leading to significant resource consumption associated with the autoscaling system. To address this, we propose GraphOpticon, a novel proactive horizontal autoscaling framework that leverages a singular global forecasting model based on Spatiotemporal Graph Neural Networks. The experimental results demonstrate that GraphOpticon is capable of providing improved service performance, and resource consumption (caused by the workloads involved and the autoscaling system itself). As a matter of fact, GraphOpticon manages to consistently outperform other contemporary horizontal autoscaling solutions, such as Kubernetes’ Horizontal Pod Autoscaler, with improvements of 6.62% in median execution time, 7.62% in tail latency, and 6.77% in resource consumption, among others.

The Edge-Cloud Continuum is a large-scale, loosely coupled system consisting of multiple stakeholders, regions, dynamic infrastructures, and conflicting objectives. With surging growth and demand, the Continuum’s energy and carbon footprint have massively increased, resulting in great operational expense, environmental impact, and strain on power grids. Methods to mitigate this face significant challenges: Quality of Service (QoS) guarantees must be balanced against not only carbon emissions, but the loadings, capacities, and QoS of the (smart) grids that power the underlying infrastructure. Integrated models to enable reasoning across both a Continuum and its associated SmartGrids are therefore required.

This work presents a formal model to reason across the integration of Smart Grids and the Edge-Cloud Continuum. Firstly, we identify the components, interactions, and properties crucial to mitigating cross-Continuum energy and carbon footprint while maintaining user, provider, and power grid QoS. We then present associated mathematical models to enable a model-based simulation to be developed based on our work. We present this simulation (all code is available for download) and use a simple scheduling algorithm to demonstrate the feasibility of utilizing knowledge from both the Smart Grid and EdgeCloud Continuum for carbon and energy management, showing that significant savings are possible

Compute clusters are major power consumers in Cloud and Edge data centers, making it critical to reduce power usage and costs without compromising service levels objectives. Energy Performance Preference (EPP) settings and CPU frequency scaling can lower power but typically at the cost of reduced performance. When considering clusters with heterogeneous power profiles, it is essential to map workloads to the most suitable profile based on their quality-of-service constraints. Current orchestrators overlook power-profile heterogeneity; this is a particular concern at the Edge, where otherwise identical hardware may range from power-optimized to performance-oriented yet remain indistinguishable to schedulers. We present a taxonomy of power-aware orchestration, and extend the default Kubernetes scheduler with power-profile awareness. We evaluate the feasibility of this extended scheduler by comparing three power profile-aware scheduling strategies on a testbed running a microservices benchmark, with results showing that average power use can be reduced by up to 12% while maintaining application performance. We conclude with key challenges and future research directions.

In the rapidly evolving landscape of Software-Defined Networking (SDN), the enhancement of security measures against sophisticated cyber threats is paramount. Among these threats, inference attacks pose a significant risk by allowing adversaries to deduce the configurations and policies of SDN switches, thereby undermining the integrity and confidentiality of the network infrastructure. To address this critical issue, we introduce a novel dynamic rule replacement policy for SDN switches, leveraging the capabilities of a Support Vector Machine (SVM) for its implementation. Our approach utilizes a comprehensive set of statistical features, including duration analysis of flow rules, dispersion of packet match fields, and frequency of packet arrivals to identify patterns indicative of potential inference attacks. By dynamically adjusting the rules within SDN switches based on the analysis of these features, our policy significantly enhances the resilience of the network against such attacks. To accelerate the innovation and development of network services, this study proposes an integrated SDN architecture deployed over a serverless framework. This work serves as a starting point to enable researchers to realize the concept of modular serverless functions over traditional SDN environments. We show during inference attacks how a serverless framework improves the latency and resource utilization of the network compared to a traditional SDN framework. This study demonstrates an improvement in preventing inference attacks without compromising the performance and efficiency of the SDN infrastructure.

Cloud infrastructures are evolving from centralised systems to geographically distributed federations of edge devices, fog nodes, and clouds - often known as the Cloud-Edge Continuum. Continuum systems are dynamic, often massive in scale, and feature disparate infrastructure providers and platforms; this greatly increase the complexity of developing and managing applications. The Serverless paradigm shows the potential to greatly simplify the process of building Continuum applications - however, current scheduling mechanisms for Serverless Continuum platforms pay little attention to reducing the energy consumption and improving the sustainability of function execution. This is a significant omission, made worse as computing nodes within a Continuum may be powered by renewable energy sources that are intermittent and unpredictable, making low-powered and bottleneck nodes unavailable.There is great opportunity to design a decentralized energy management scheme for scheduling Serverless functions that takes advantage of the different layers of the Continuum, such as IoT devices located at the Edge, on-premises clusters closer to the data sources, or directly on large Cloud infrastructures. To achieve this, we formally model a green energy-aware Serverless workload scheduling problem for the multi-provider Cloud-Edge Continuum. We then design stable matching based technique for decentralized energy management (utilising a distributed controller) that considers the availability of green energy nodes and the QoS requirements of Serverless functions. We prove the complexity, stability and termination of the proposed heuristic algorithm, and also compare its performance with baseline scheduling techniques.

We propose a novelty system to manage Traffic Priority at city intersections by means of our Mobility-Hub (M-Hub), a next-generation Traffic Light Controller that leverages the power of cloud-edge continuum computing, Digital Twin, and Cellular Vehicle-to-Everything (C-V2X) technologies to transform traffic management into a dynamic and intelligent system. M-Hub acts as an open-edge computing platform, enabling real-time data processing, 3rd parties containerized applications and decision-making at the network edge. COGNIT is an open-source cloud-edge continuum framework, that offers many improvements for next-generation Intelligent Transportation Systems (ITS). The continuum allows for the integration of diverse data sources, including vehicular data from C-V2X communication, real-time traffic information from detectors or cameras, and other environmental data, to seamlessly generate Digital Twins in the ACISA smart mobility platform, SATURNO. By combining this data with advanced traffic optimization algorithms implemented in the COGNIT infrastructure, M-Hub can dynamically adjust traffic signal timings, optimize traffic flow, and reduce congestion with optimal use of computational resources. M-Hub has the potential to revolutionize urban mobility, enhancing safety, improving efficiency, and reducing environmental impact.