Worldwide, network virtualization has attracted attention as a promising network technology and is expected to be utilized in new-generation networks. In this paper, we propose a topology design and admission control approach for such network virtualization with improved network robustness of both the physical and virtual networks. In our proposed method, first we construct a new network graph that includes the performance of the nodes. We construct a new Laplacian matrix to evaluate the network robustness of the physical networks. Then, based on a user's request, a topology of a virtual network is determined with the KMB algorithm. This topology design is represented as a minimum Steiner tree problem. Moreover, by using admission control, the virtual network is created so as not to degrade the network robustness of the physical networks. If the network robustness of physical networks is degraded, the user's request is rejected and the virtual network is not created. With a simulation, we evaluated the performance of our proposed topology design and admission control approach. Numerical examples show that by using our proposed method, a large number of virtual networks can be provided for users. Moreover, the network robustness of the virtual networks can be improved. Our proposed method is effective for improving the network robustness of both the physical and virtual networks.
In this paper, in order to embed virtual networks by considering network security, we propose a virtual network embedding based on security level with VNF placement. In this method, virtual networks are embedded in a substrate network by considering security and some security VNFs are placed in order to increase the security level of substrate networks. By using our proposed method, many virtual networks can be embedded by considering security level. As a result, the reward can be increased and the cost of placing VNFs is not increased so much. We evaluate the performance of our proposed method with simulation. The performance of this method is compared with the performance of a method that places VNFs randomly and the performance of a method without placing VNFs. From numerical examples, we investigate the effectiveness of this method. In numerical examples, we show that the proposed method is effective in embedding virtual networks by considering network security.
In network virtualization, it has been considered that virtual networks are constructed over a physical network where conventional data transmission services have been utilized. Here, virtual networks have to be constructed while keeping qualities of the conventional services. In this paper, we propose a new virtual network construction in order to construct many virtual networks while keeping the robustness of a physical network by using network resources effectively. The proposed method consists of three processes: K-shortest path algorithm and Prim's minimum spanning tree algorithm, path splitting, and path migration. In the proposed method, at first, multiple topologies are designed by using the K-shortest path algorithm and the Prim's MST algorithm according to the user's request. After the topology design is completed, an admission control with network robustness of the physical network is performed. Then, if one of the designed topologies can satisfy the construction conditions, a virtual network is constructed and provided with the user. Otherwise, the path splitting and path migration are performed. Here, the path splitting is utilized to design another topology of a virtual network and path migration is used to change the topologies of the virtual networks that have already been constructed. These processes are formulated as optimization problems and those are processed by solving the optimization problems. In numerical examples, we show that our proposed method can construct a higher number of virtual networks while keeping the robustness of a physical network by comparing with the conventional method where only the Kou-Markowsky-Berman algorithm is used. Int. J. Commun. Syst. 2017; 30: e2958 1 of 21 Y. URAYAMA AND T. TACHIBANA flow table that describes each action for each data packet. Data packets can be forwarded to a corresponding virtual network according to the flow table in the OpenFlow controller.Here, the amount of available network resources is limited in the physical network, and hence, each virtual network has to be constructed so that the network resources are utilized efficiently. Some methods for designing the topology of a virtual network and allocating the network resources to the virtual network have been proposed in the literatures.In [10], in order to construct a virtual network simply over a large physical network, a topology of the virtual network is designed by combining two or more small star topologies. Yu et al. [11] has proposed a construction method of a virtual network with path splitting and path migration.In this method, a service provider obtains the revenue from a user after a virtual network was constructed. In order to maximize the revenue, virtual networks are constructed over a physical network by using the path splitting and path migration.In particular, by using the path migration, the topologies of constructed virtual networks are redesigned periodically. In [12], an architecture called DaVinci has been proposed so that network resources can be allocated to each ...
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.