Design of survivable IP-over-optical networks

Abstract

In the past years, telecommunications networks have seen an important evolution with the advances in optical technologies and the explosive growth of the Internet. Several optical systems allow a very large transport capacity, and data traffic has dramatically increased. Telecommunications networks are now moving towards a model of high-speed routers interconnected by intelligent optical core networks. Moreover, there is a general consensus that the control plan of the optical networks should utilize IP-based protocols for dynamic provisioning and restoration of lightpaths. The interaction of the IP routers with the optical core networks permits to achieve end-to-end connections, and the lightpaths of the optical networks define the topology of the IP network. This new infrastructure has to be sufficiently survivable, so that network services can be restored in the event of a catastrophic failure. In this paper we consider a multilayer survivable network design problem that may be of practical interest for IP-over-optical neworks. We give an integer programming formulation for this problem and discuss the associated polytope. We describe some valid inequalities and study when these are facet defining. We discuss separation algorithms for these inequalities S. Borne ( ) Springer 42 Ann Oper Res (2006) 146:41-73 and introduce some reduction operations. We develop a Branch-and-Cut algorithm based on these results and present extensive computational results.

Genetic Algorithm for Solving Multilayer Survivable Optical Network Design Problem

“…In 2006, S. Borne et.al studied survivable IP-over-optical network design problem [2]. They gave a 0-1 integer pro- gramming formulation for this problem, described some valid inequalities and discussed separation algorithms for these inequalities, simultaneously, introduced some reduction operations.…”

Solving survivable two-layer network design problems by metric inequalities

“…where (7) and (8) In fact, even the simple assumption of choosing in advance physical paths for the lightpaths, as we do in (EF F ) and (ECF ), is a routing constraint that can not be handled by metric inequalities (8). Formulation (ICF ) can be used only under the assumption that the physical paths for the [24]).…”

“…Because optical transport networks are designed to carry high volumes of traffic, network failures may have severe consequences. Therefore survivability requirements have became an integral part of every network design problem [8], [16], [29]. As a result, the complexity of the design and configuration process increases significantly.…”

“…If the design problem is solved making sure that the capacity constraints on the physical links are satisfied, it is always possible to compute a feasible assignment of wavelengths to the lightpaths placing wavelength converters, if needed. In this paper, following [3], [4], [8], [10], [6], [25], [18] and others, we focus our attention on the network design problem. We refer the reader to [19], [20] and references therein for the problem of installing wavelength converters in the network at minimum cost.…”

“…In [25] MIP-based heuristics to be used within a branch-and-cut scheme are presented. In [8], valid inequalities and a branch-and-cut approach are presented for the uncapacitated case with survivability requirements. In [13] multi-layer hop-constrained node-survivable networks are addressed and two different survivability mechanisms (path diversity and path protection) are considered.…”