Littérature scientifique sur le sujet « Reconfiguration des VNFs »

Satellite–terrestrial integrated networks (STINs) are regarded as a promising solution to meeting the demands of global high-speed seamless network access in the future. Software-defined networking and network function virtualization (SDN/NFV) are two complementary technologies that can be used to ensure that the heterogeneous resources in STINs can be easily managed and deployed. Considering the dual mobility of satellites and ubiquitous users, along with the dynamic requirements of user requests and network resource states, it is challenging to maintain service continuity and high QoE performance in STINs. Thus, we investigate the service migration and reconfiguration scheme, which are of great significance to the guarantee of continuous service provisioning. Specifically, this paper proposes a dynamic service reconfiguration method that can support flexible service configurations on integrated networks, including LEO satellites and ground nodes. We first model the migration cost as an extra delay incurred by service migration and reconfiguration and then formulate the selection processes of the location and migration paths of virtual network functions (VNFs) as an integer linear programming (ILP) optimization problem. Then, we propose a fuzzy logic and quantum genetic algorithm (FQGA) to obtain an approximate optimal solution that can accelerate the solving process efficiently with the benefits of the high-performance computing capacity of QGA. The simulation results validate the effectiveness and improved performance of the scheme proposed in this paper.

In Network Function Virtualization, the resource demand of the network service evolves with the change of network traffic. VNF dynamic migration has become an effective method to improve network performance. However, for the time-varying resource demand, how to minimize the long-term energy consumption of the network while guaranteeing the Service Level Agreement (SLA) is the key issue that lacks previous research. To tackle this dilemma, this paper proposes an energy-efficient reconfiguration algorithm for VNF based on short-term resource requirement prediction (RP-EDM). Our algorithm uses LSTM to predict VNF resource requirements in advance to eliminate the lag of dynamic migration and determines the timing of migration. RP-EDM eliminates SLA violations by performing VNF separation on potentially overloaded servers and consolidates low-load servers timely to save energy. Meanwhile, we consider the power consumption of servers when booting up, which is existing objectively, to avoid switching on/off the server frequently. The simulation results suggest that RP-EDM has a good performance and stability under machine learning models with different accuracy. Moreover, our algorithm increases the total service traffic by about 15% while ensuring a low SLA interruption rate. The total energy cost is reduced by more than 20% compared with the existing algorithms.

Ces dernières années, en raison de la croissance sans précédent du nombre d'appareils connectés et de données mobiles, et des développements continus des technologies pour répondre à cette énorme demande de données, le réseau de cinquième génération (5G) a émergé. La future architecture 5G sera essentiellement basée sur le Network Slicing (NS), qui permet de fournir une approche flexible pour réaliser la vision 5G. Grâce au concept émergent de virtualisation des fonctions réseau (NFV), les fonctions réseau sont découplées des matériels physiques dédiés et réalisées sous forme de logiciel. Cela offre plus de flexibilité et d'agilité dans les opérations commerciales. Malgré les avantages qu'il apporte, NFV soulève quelques défis techniques, le problème de reconfiguration étant l'un d'entre eux. Ce problème, qui est NP-difficile, consiste à réaffecter les fonctions de réseau virtuel (VNFs) pour s'adapter aux changements du réseau, en transformant l'état courant des services déployés, on peut illustrer cela par la migration des machines virtuelles (VM) qui hébergent les VNF, à un autre état qui répond aux objectives des opérateurs. Cette thèse de doctorat étudie comment reconfigurer les VNFs en les migrant vers un état optimal qui pourrait être calculé en avance ou inconnu. Dans cette thèse, nous avons étudié les deux cas en minimisant la durée d'interruption de service et la durée de migration des VNFs. Nous avons proposé des méthodes exactes et approchées. Parmi les méthodes exactes, nous citons deux modèles PLNE. Nous avons également proposé deux approches heuristiques, l'une basée sur la génération de colonnes et la deuxième utilisant la notion de “feedback arc set". L'objectif global de ce travail est donc de définir et d'étudier le problème de reconfiguration des VNFs dans le contexte du 5G network slicing, et de proposer des modèles mathématiques et des algorithmes efficaces pour résoudre les problèmes d'optimisation sous-jacents
In recent years, because of the unprecedented growth in the number of connected devices and mobile data, and the ongoing developments in technologies to address this enormous data demand, the fifth generation (5G) network has emerged. The forthcoming 5G architecture will be essentially based on Network Slicing (NS), which enables provide a flexible approach to realize the 5G vision. Thanks to the emerging Network Function Virtualization (NFV) concept, the network functions are decoupled from dedicated hardware devices and realized in the form of software. This offers more flexibility and agility in business operations. Despite the advantages it brings, NFV raises some technical challenges, the reconfiguration problem is one of them. This problem, which is NP-Hard, consists in reallocating the Virtual Network Functions (VNFs) to fit the network changes, by transforming the current state of deployed services, e.g., the current placement of Virtual Machines (VM) that host VNFs, to another state that updates providers’ objectives. This PhD thesis investigates how to reconfigure the VNFs by migrating them to an optimal state that could be computed in advance or free placement. In this thesis, we studied both cases while minimizing the service interruption duration and the VNF migration duration. We have proposed exact and approximate methods. Among the exact methods, we cite two ILP models. We also proposed two heuristic approaches, one based on column generation and the second using the concept of “arc set feedback”. The overall objective of this work is therefore to define and study the problem of VNF reconfiguration problem in the context of 5G network slicing, and propose mathematical models and efficient algorithms to solve the underlying optimization problems