With the fierce global competition on satellite networks, the building of satellite constellations grows explosively. Sharply increasing on-orbit data will face the challenge of satellite-ground data transmission. GEO satellites become the top choice for satellite data relay due to their stable satellite-ground link. Most existing spectrum resource management for GEO relays is equipment-oriented and benefit priority, which may lead to a waste of spectrum resources. In this paper, we propose a real-time task-oriented resource allocation strategy for GEO relay systems. We model the spectrum allocation problem as a distributed non-cooperative Stackelberg game process. We prove that when both sides of the game pursue the maximization of personal revenue, the system will enter a Nash equilibrium state, whereas spectrum resources are not fully used. Based on the maximization of individual utilities (U-prior) and spectrum utilization (S-prior) methods, we design an adaptive dual-mode pricing mode to maximize the spectrum resources within a certain loss of revenue. The simulation results show that the S-prior and U-prior have better performance than the baseline method and existing optimization methods. Our proposed dual-mode strategy is making more throughputs and has less delay with little loss of utility values than that of individual utility maximization.

Network softwarization makes it easy to quickly deploy various and different network services with the composition of virtual network functions (VNFs) that can be launched, modified and stopped at any time. Nowadays, several VNF servers supporting different types of VNFs exist. Therefore, to minimize the cost of deploying network services, VNFs should not only be placed on the best servers but also chained in an optimal way. In this paper, we propose a new approach to solve the NP-hard VNF placement and chaining problem (VNFPC problem). After proving that VNFPC problem can be transformed to a variant of the multi-constrained routing problem where the number of additive metrics is part of the problem, we proposed efficient heuristic reducing the worst-case time complexity while ensuring high quality solutions. Simulation results show that our constrained shortest paths-based heuristics allows to determine solutions close to the optima by keeping small number of paths on nodes.

During the last decade, we are witnessing the emergence of NFV and SDN to reduce CAPEX and OPEX. Under the SDN paradigm and thanks to NFV, a service can be swiftly deployed by the chaining of several VNFs forming an SFC running on a virtualized infrastructure. Nowadays, there are still quite a number of issues related to SFCs, among them, the optimal placement of SFC components. In this paper, we focused on the variant of the resource allocation cost optimization problem of VNF placement and chaining for limited resources on the servers. After proving that the problem of VNF placement is NP-Hard and equivalent to the multiple knapsack problem, we proposed a genetic algorithm-based meta-heuristic to solve large instance of our VNF placement and chaining problem variant. Simulation results show that our genetic algorithms are efficient since they reduce the SFC mean cost and improve the accepted requests ratio.

Network Function Virtualization (NFV), which decouples network functions from hardware and transforms them into hardware-independent Virtual Network Functions (VNF), is a crucial technology for many emerging networking domains, such as 5G, edge computing and data-center network. Service Function Chaining (SFC) is the ordered set of VNFs. The VNF deployment problem is to find the optimal deployment strategy of VNFs in SFC while guaranteeing the Service-Level Agreements (SLAs). Existing VNF deployment researches mainly focus on sequences of VNFs without energy consideration. However, with the rapid development of application requirement, the SFCs evolve from sequence to dynamic graph and the service providers become more and more sensitive to the energy consumption in NFV. Therefore, in this paper, we identify the Energy-efficient Graph-structured SFC problem (EG-SFC) and formulate it as a Combinatorial Optimization Problem (COP). Benefiting from the recent advances in machine learning for COP, we propose an end-to-end Graph Neural Network (GNN) based on constrained Deep Reinforcement Learning (DRL) method to solve EG-SFC. Our method leverages the Graph Convolutional Network (GCN) to represent the Q-network of Double Deep Q-Network (DDQN) in DRL. The mask mechanism is proposed to deal with the resources constraints in COP. The experimental results show that the proposed method can deal with unseen SFC graphs and achieve better performances than greedy algorithm and traditional DDQN.

Network virtualization allows the co-existing of logical networks (virtual networks) on physical network (substrate networks). Virtual Network (VN) reliability is a critical problem for end-users and service providers. It aims to ensure service continuity even upon failure. As more and more VNs are created over substrate networks (SN), the failure of a single SN component may lead to the failure of many VNs. Thus, the VN reliability issue is becoming more and more critical. VN reliability can be enhanced in two ways: (1) by failure recovery (post-failure) with protection and/or restoration methods; (2) by failure avoidance with the selection of most reliable components at the network topology setting phase. Traditional virtual network embedding (VNE) methods have mainly focused on bandwidth optimization. In this paper, we focus on the reliability issue. We propose VNE methods which take into account the failure probability of SN components with a failure-avoidance approach, in order to minimize the VN failure probability. Our heuristics are based on the use of Steiner Minimal Tree (SMT). Simulations results confirm that our heuristics provide better reliability against traditional VNE with bandwidth as sole target, and, in case of failure of a SN component, reduce the number of affected VNs.