Effective resource management in edge-cloud networks requires accurate prediction of resource utilization across diverse workloads. The dynamic nature of user demands necessitates prediction models with strong generalization capabilities, i.e., able to achieve high performance in sudden traffic changes or even unseen patterns. Existing works struggle with long-term dependencies and varied temporal patterns. This paper proposes OmniFORE (Framework for Optimization of Resource forecasts in Edge-cloud networks). In particular, we combine attention-based time-series models with temporal clustering to achieve robust generalization and efficiently consume and predict diverse workloads in volatile environments. By training on representative subsets from extensive datasets, OmniFORE captures both short-term stability and long-term changes in resource usage patterns. Experiments with real-world data show that our approach outperforms state-of-the-art methods, particularly in new environments, improving prediction accuracy by 83.87% and achieving 17.92% faster inference.

The sixth generation (6G) of mobile networks seeks to leverage Integrated Terrestrial and Non-Terrestrial Networks (ITNTN) to support next-generation applications. However, despite increased network densification, meeting the stringent latency requirements—particularly for delay-sensitive applications—remains a significant challenge due to the high latency overhead introduced by non-terrestrial nodes. Content caching offers a potential solution to reduce content access delays and backhaul bandwidth costs by pre-storing content closer to the end users. However, the heterogenity of nodes, the varying content popularity and the distinct Quality of Service (QoS) requirements introduced by network slicing complicate caching decisions in ITNTNs. For instance, prioritizing content related to critical services (e.g., emergency services) over popular content (e.g., social media files) may better address QoS needs, albeit with lower hit ratios, an aspect not explored in existing works. This paper introduces SaCCA-RL (Reinforcement Learning-based Slice-aware Content Caching Algorithm) for selecting which content to cache and its placement under constrained ITNTN node capacities, considering different and varying content popularity and priority. Simulations show that SaCCA-RL results in more than 10% improvement in terms of cache gain compared to baseline algorithms, while guaranteeing 100% hit-ratio for critical content.

The virtualization and softwarization of 5G/6G mobile networks have enabled the deployment and orchestration of cloud-native network and application functions. The deployment of these functions is crucial, as the placement of data plane elements (i.e., User Plane Function (UPF)) and vertical services can significantly impact the overall user latency. However, in multi-slice edge scenarios, characterized by users with distinct levels of criticality, the problem of UPF and application placement is becoming increasingly complex due to i) the various costs involved and ii) the limited computational resources at the edge. In this paper, the problem of joint UPF and application placement for a multi-slice user scenario is studied, taking into account multiple cost components that influence the placement decision, including service migration, traffic forwarding, server activation and processing costs. To tackle this problem, we introduce a Joint UPF and Application Reinforcement Learning-based (JUAP-RL) algorithm, which decides the UPF and application deployment location and coordinates the placement stages. Extensive experiments have shown that JUAP-RL demonstrates up to 17% gain in terms of user acceptance ratio and up to 23.4% reduction in provisioning cost compared to baseline schemes.

Future applications to be supported by 6G networks are envisaged to be realized by loosely-coupled and independent microservices. In order to achieve an optimal deployment of applications, smart resource management strategies will be required, working in a cost-effective and resource-efficient manner. Current cloud computing services are challenged to meet the explosive growth and demand of future use cases such as virtual/augmented/mixed reality (VR/AR/MR). The purpose of edge computing (EC) is to better address latency and transmission requirements of those future stringent applications. However, a high flexibility and a rapid decision-making will be required since EC suffers from limited resources availability. For this reason, this work proposes an artificial intelligence (AI) technique, based on reinforcement learning (RL), to make intelligent decisions on the optimal tier and edge-site selection to serve any request according to the application’s category, constraints, and conflicting costs. In addition, when deployed at the edge-network, a heuristic has been proposed for the mapping of microservices within the selected edge-site. That heuristic will exploit a ranking methodology based on the network topology and available network and compute resources while preserving the revenue of the mobile network operator (MNO). Simulation results show that the performance of the proposed RL approach is close to the optimal solution by reaching the cost minimization objective within a 8.3% margin; moreover, RL outperforms considered benchmark algorithms in most of the conducted experiments.

Cognizant of the ease with which softwarized functions can be dynamically scaled according to real time resource requirements, and the fact that multiple services can have common VNFs in their chaining, this paper tackles the problem of cost effective deployment of online services from the perspective of sharing their VNF instances. First, we formally formulate the deployment problem under VNFs sharing. Secondly, given the NP-hard nature of the above problem, we propose a reinforcement learning (RL) algorithm capable of making intelligent placement decisions while considering multiple conflicting costs. Costs of transmission, VNF instantiation or energy consumption, among others. Thanks to the intelligence of the RL algorithm, simulation results show that the performance of the proposed algorithm is within a 14% margin and similar to an optimal solution in terms of request provisioning cost and acceptance ratio, respectively. Moreover, the algorithm results in more than a 20% and a 70% improvement in terms of request deployment cost and time compared to a state-of-the-art algorithm, and up to more than a 40% improvement in terms of cost compared to an algorithm that greedily minimizes the transmission or VNF activation costs.

Network Function Virtualization (NFV) and Machine Learning (ML) are envisioned as possible techniques for the realization of a flexible and adaptive 5G network. ML will provide the network with experiential intelligence to forecast, adapt and recover from temporal network fluctuations. On the other hand, NFV will enable the deployment of slice instances meeting specific service requirements. Moreover, a single slice instance may require to be deployed across multiple substrate networks; however, existing works on multi-substrate Virtual Network Embedding fall short on addressing the realistic slice constraints such as delay, location, etc., hence they are not suited for applications transcending multiple domains. In this paper, we address the multi-substrate slicing problem in a coordinated manner, and we propose a Reinforcement Learning (RL) algorithm for partitioning the slice request to the different candidate substrate networks. Moreover, we consider realistic slice constraints such as delay, location, etc. Simulation results show that the RL approach results into a performance comparable to the combinatorial solution, with more than 99% of time saving for the processing of each request.