p-Cycles: New Solutions for Node Protection, Transparency, and Large Scale Network Design

Optical transport networks are critical infrastructures which carry huge amounts of great traffic variety but are inevitably subject to laser diode failures, fiber cuts and sometimes node outages. The concept of p -cycles is very attractive and competitive in the domain of network survivability because p-cycles have a unique ability to combine the real-time switching simplicity and speed of rings with the capacity efficiency, flexibility and freedom of a mesh in the routing of working and restored state paths.;This dissertation presents several new research studies that increase our knowledge and act of available techniques to use and understand p-cycles. Advancements include a relatively simple but cost-effective generalization of how a BLSR-ring (or p-cycle to-date) derives survivability, in the event of node failure, through loopback at the nearest two neighbor-nodes on the same cycle. Significantly, this new insight also gives rise to a novel two-hop segment protection paradigm that unifies node and span failure protection.;As well, the thesis introduces two fundamental advances for dealing with optical network transparency. One is the complementary matching of longer working paths with shorter protection segments available through p-cycles, thereby controlling the optical reach in restored network states. The other is the in-depth consideration of glass-switched p-cycles to rapidly, simply and efficiently provide for the direct replacement of failed fiber sections with whole replacement fibers. Experiments highlight that p-cycles formed out the span fibers overcome the complexity due to wavelength continuity requirements in transparent-based designs, significantly reduce overall capital expenditure (CapEx) costs and provide a solid working capacity envelope for dynamic traffic considerations.;We also make advances on the problem of solving very large scale p-cycle design problems, with a technique that combines Genetic Algorithms (GA) with Integer Linear Programming (ILP). Basically, the GA-ILP considers any p-cycle ILP (to be solved) as the fitness function for a GA-like evolutionary heuristic, aimed at preselecting a few manageable candidate cycles working well together, from an almost infinite space. Beside the GA-ILP conceptual simplicity, experiments show high quality design solutions to very large scale p-cycle problem instances, involving up to 200 nodes.