Creating and sharing knowledge for telecommunications

A Re-Optimization Approach for Virtual Network Embedding

Melo, M. ; Carapinha, J. ; Sargento, S. ; Killat, U. Killat ; Timm-Giel, A.

A Re-Optimization Approach for Virtual Network Embedding, Proc ICST Conf. on Mobile Networks and Management (MONAMI), Hamburg, Germany, Vol. N/A, pp. N/A - N/A, September, 2012.

Digital Object Identifier: 10.1007/978-3-642-37935-2_21

Download Full text PDF ( 528 KBs)

Abstract
Network Virtualization is claimed to be a key component of the Future Internet by enabling the coexistence of heterogeneous (virtual) networks on the same physical infrastructure, providing the dynamic creation and support of different networks with different paradigms and mechanisms in the same physical network. A major challenge in the dynamic provision of virtual networks resides on the efficient embedding of virtual resources into physical ones.
Since this problem is known to be NP-hard, previous research focused on designing heuristic-based algorithms; most of them do not consider either a simultaneous optimization of the node and the link mapping or the re-optimization of VNs, leading to non-optimal solutions.
This paper proposes an extension to VNE-NLF in order to support the re-optimization of existing VNs on the substrate and to provide the optimal boundary. In addition, an evaluation is done on the methods VNE-NLF and VNE-ESPH to assess the impact on the overall performance due to either the VN size or to the physical network size. Simulation experiments show significant improvements when using the VN re-optimization process. Not only the bandwidth consumption had been smaller reaching values of minus 17.5%, but it had also reduced the maximum utilization levels on the CPU and on the memory, where it achieves maximum levels of utilization of minus 20% and minus 16%, respectively. The VN size do not reveal to affect the performance of the VNE-NLF, although the same do not same to happen to the VNE-ESPH, where it loses performance with the size of the VN. Regarding the physical network size both methods seem to be affected by it, where larger physical networks accept more VNs.