5G networks promise to deliver an unprecedented performance that can accommodate novel services with stringent Quality of Service (QoS) requirements that were not possible with previous generations of networks. Edge Computing plays a fundamental role by providing computing resources closer to the user, reducing round trip times. However, the deployment of edge computing poses new challenges, including the energy footprint of a potentially large number of servers. Even in idle state, these servers consume a significant amount of energy, which is worth considering for reducing their energy footprint. In cloud computing environments, server shutdown during low-demand periods is a typical energy-saving strategy. However, this approach has received less attention in edge computing due to the strict latency requirements of its use cases. This work presents ODESA, an edge server shutdown strategy with polynomial time complexity that provides a tradeoff between the idle energy consumption of the edge servers and energy consumed by the backhaul to route requests to active servers. Our numerical investigation shows that thanks to the reduction in idle energy consumption, ODESA reduces the total consumption by 42% over the common always-on approach during low-demand periods and 11% over 24 hours, all while meeting the latency requirements of the applications.

While advances in wireless networks enable novel services with previously unreachable latency guarantees, edge computing becomes essential for delivering computing resources close to the users and meeting the strict latency requirements. However, addressing the energy footprint of computing resources is crucial amid the pressing sustainability concerns. The energy consumption of idle resources accounts for a significant part of the total energy footprint. While server shutdown during low-demand periods is common in cloud computing, it is challenging to determine which edge servers to shut down and how to route requests due to the stringent latency requirements of the applications. Thus, this work formulates an optimal orchestration policy to minimize the energy consumption of the edge computing infrastructure and presents LESS-ON, a strategy with a polynomial time complexity that reduces the operational energy footprint of edge computing by shutting down edge servers during low-demand periods. In contrast to previous studies, LESS-ON considers the energy requirements associated with routing requests to the designated edge servers. Our numerical evaluation shows that LESS-ON reduces the total consumption by 42% with respect to the common always-on approach during low-demand periods and by 35% over 24 h, all while meeting latency requirements.

Abstract While cloud computing is crucial in processing data from devices with low computational power, the latency introduced by the Internet backhaul limits real-time applications. By situating computing resources at the network’s edge, edge computing offers low-latency services by offloading computations from high-performance computing (HPC) data centers to the edge servers, reducing wide Area network (WAN) strain. As a result, edge computing has unlocked opportunities for innovative applications that were previously unfeasible, such as connected vehicles or medical robotics. Nonetheless, deploying the infrastructure required to support edge computing services raises sustainability and energy consumption concerns. Consequently, the development of tools enabling researchers to explore innovative approaches to reducing the energy impact of edge computing is crucial. In this work, we present MintEDGE, a network simulator focused on the energy consumption of edge computing. Our simulator allows testing energy-saving approaches and task placement algorithms in realistic large-scale scenarios encompassing entire regions.

Edge computing has transformed cellular networks, offering fast response times by moving computing resources to the network's edge. This not only reduces the burden on the Wide Area Network (WAN) but also enables latency-sensitive applications. However, the widespread deployment of edge computing raises concerns regarding its sustainability. In this work, we present MintEDGE, a simulation framework that models a fully configurable edge-enabled cellular network. MintEDGE empowers researchers and practitioners to design and assess energy-saving strategies for edge computing. We discuss the details of the simulator and its customizable elements like user mobility, the possibility to use predictive workload algorithms, and diverse application scenarios at scale. MintEDGE is released under a permissive MIT license.

With the forthcoming of 5G networks, the underlying infrastructure needs to support a higher number of heterogeneous services with different QoS needs than ever. For that reason, 5G inherently provides a way to allocate these services over the same infrastructure through the concept of Network Slicing. However, to maximize revenue and reduce operational costs, a method to proactively adapt the resources assigned to each slice becomes imperative. For that reason, this work presents two Machine Learning (ML) models, leveraging Long-Short Term Memory (LSTM) and Random Forest algorithms, to forecast the throughput of each slice and adapt accordingly the amount of resources needed. The models are evaluated using NS-3, which has been integrated with the ML models through a shared memory framework. This enables a closed loop in which the predictions of the models can be used at run time to introduce changes in the network. Consequently, it makes it able to cope with the forecasted requirements, eliminating the need for off-line training and resembling better a real-life scenario. The evaluation performed shows the ability of the models to predict the slices' throughput under various settings and proves that Random Forest provides up to 26% better results than LSTM.

The COVID-19 emergency has made the consumption of multimedia content skyrocket in all contexts, including education. Many universities leverage hybrid learning models, in which students join a real-time video session via Wi-Fi from several classrooms to ensure safety and social distancing. This is creating a significant strain on the wireless access network, which is required to deliver an unusually high level of traffic. Artificial Intelligence (AI) and Machine Learning (ML) solutions have emerged as a way to make networks easier to control and to manage. However, their black box nature and in general their fire and forget approach has generated considerable skepticism over the entire value chain, from vendors to network administrators. This situation has led to a new interest in interpretable AI solutions, which aim at making the decisions taken by AI/ML models intelligible to a domain expert. In this article, we review the concept of interpretable AI and analyze the challenges, requirements, and benefits it can bring to delay-sensitive content delivery in 802.11 Wi-Fi networks. Furthermore, we apply these requirements to a use case in which we focus on advanced Quality of Service (QoS) provision, and we propose an interpretable and low-complexity ML model that addresses those requirements. The results demonstrate performance gains up to 60% in the sensitive traffic and up to 20% at network-wide level.

Although 4G and 5G Radio Access Technologies(RATs) aim to usher in faster connectivity that is able to cope with mobile traffic demands, this capability is sometimes hindered by poor indoor signal quality caused by distance from base stations and the materials used in the construction of buildings. These factors have led to Wi-Fi being adopted as the technology of choice in indoor scenarios. Although the deployment of Wi-Fi Access Points (APs) can be planned, the user-AP association procedure is not defined by the standard but left to the vendor's choice, which for simplicity is usually driven by signal strength. This approach leads to uneven user distributions and poor resource utilization. To overcome this rigidity, in this paper, we leverage SoftwareDefined Networking (SDN) to propose ajoint user association and channel assignment solution in Wi-Fi networks. Our approach considers average signal strength, channel occupancy, and AP load to make better user association decisions. Experimental results have demonstrated that the proposed solution improves the aggregated goodput by 22% with respect to approaches based on signal strength. Furthermore, user level fairness is also improved.