Open AccessShared end-to-content backup path protection in k-node (edge) content connected elastic optical datacenter networks
Author(s) -
Xin Li,
Shanguo Huang,
Shuo Yin,
Bingli Guo,
Yongli Zhao,
Jie Zhang,
Min Zhang,
Wanyi Gu
Publication year - 2016
Publication title -
optics express
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.394
H-Index - 271
ISSN - 1094-4087
DOI - 10.1364/oe.24.009446
Subject(s) - backup , computer science , survivability , computer network , node (physics) , enhanced data rates for gsm evolution , path (computing) , path protection , content (measure theory) , heuristic , distributed computing , wavelength division multiplexing , telecommunications , optics , mathematics , wavelength , mathematical analysis , physics , structural engineering , database , artificial intelligence , engineering
To quantitatively measure content connectivity and provide protection for different kinds of content, the concept of k-node (edge) content connectivity is proposed recently. Based on k-node (edge) content connectivity, k-node (edge) content connected elastic optical datacenter network (KC-EODN) is proposed to design disaster-resilient and spectrum-efficient optical datacenter networks. In KC-EODN, k independent end-to-content paths are established for each request. However, it will consume too much resource to assign dedicated spectrum for each end-to-content path. Spectrum sharing among multiple end-to-content paths of different requests can greatly improve resource efficiency. In this paper, a novel perfect matching based sharing principle among multiple end-to-content paths of different requests is proposed. Based on the new proposed sharing principle, we present the shared end-to-content backup path protection (SEBPP) scheme for KC-EODN. Integer linear program (ILP) model and heuristic algorithms are designed for SEBPP scheme with the objective of minimizing the total of working and backup spectrum resources. Numerical results show that the proposed SEBPP scheme can greatly reduce spectrum consumption while ensuring the survivability against natural disaster and multi-failures.