Abstract
Telecommunication is a fast growing industry entailing regular technology developments. Communication service providers (CSPs) have to adapt to the latest technologies by upgrading their networks. For instance, most CSPs are expected to operate SONET/SDH infrastructures until the end of this decade (2020), before ultimately migrating to IP-centric software-defined networks (SDNs). Such a migration is a major undertaking. CSPs need effective optimization algorithms in order to predict the time and resources required for the circuit migration of large networks, which can stretch up to several years. We propose an optimization framework that builds a migration plan with minimum overall migration time or cost, i.e., that estimates the number of maintenance windows, engineers, and technicians required to perform the network migration. The network migration problem is formulated as a set of circuit migration problems, each requiring technician synchronization to minimize outages. We propose a two-phase algorithm that relies on the solution of column generation models to provide accurate migration plans for realistic size networks. Extensive numerical experiments are conducted on several network migration data instances arising from customers of Ciena.
© 2017 Optical Society of America
Full Article | PDF ArticleMore Like This
Omran Ayoub, Francesco Musumeci, Massimo Tornatore, and Achille Pattavina
J. Opt. Commun. Netw. 9(3) B12-B21 (2017)
Sai Kireet Patri, Elena Grigoreva, Wolfgang Kellerer, and Carmen Mas Machuca
J. Opt. Commun. Netw. 11(7) 371-382 (2019)
Carmen Mas Machuca, Mario Kind, Kun Wang, Koen Casier, Mozhgan Mahloo, and Jiajia Chen
J. Opt. Commun. Netw. 5(12) 1456-1466 (2013)