On the Complexity of Finding a Minimum Cycle Cover of a Graph
SIAM Journal on Computing
An Ultra-fast Shared Path Protection Scheme - Distributed Partial Information Management, Part II
ICNP '02 Proceedings of the 10th IEEE International Conference on Network Protocols
Partial Path Protection for WDM Networks: End-to-End Recovery Using Local Failure Information
ISCC '02 Proceedings of the Seventh International Symposium on Computers and Communications (ISCC'02)
Protection cycles in mesh WDM networks
IEEE Journal on Selected Areas in Communications
Survivable lightpath routing: a new approach to the design of WDM-based networks
IEEE Journal on Selected Areas in Communications
Extending the p-cycle concept to path segment protection for span and node failure recovery
IEEE Journal on Selected Areas in Communications
On the convolution of Pareto and gamma distributions
Computer Networks: The International Journal of Computer and Telecommunications Networking
| Hi-index | 0.00 |
This paper introduces GSP, a group shared protection scheme, for wavelength division multiplexing (WDM) mesh networks with dynamically arrived connection requests. Based on the (M:N)n control architecture, GSP has n mutually independent protection groups (PGs), each of which containing N shared risk link groups (SRLG) disjoint working paths protected by M protection paths. Due to the SRLG-disjointedness of the working paths in each PG, GSP not only allows the spare capacity to be totally sharable among the corresponding working paths, but also reduces the number of working paths affected due to a single link failure. Based on the framework, an integer linear program (ILP) formulation that can optimally reconfigure the spare capacity for a specific PG whenever a working-protection path-pair joins is proposed. This approach is appropriate for a dynamic traffic scenario where inter-arrival time is large and where arriving request can tolerate some delay, but may not be suitable where traffic arrival rate is high and incoming requests need to be served within a few seconds. To trade the performance (i.e., capacity efficiency) with the computation complexity, two heuristics, namely ring-shared protection (RSP) and link-shared protection (LSP) are proposed. The proposed schemes are compared with an exiting one, namely the successive survivable routing (SSR). The simulation results show that LSP, RSP and SSR yield similar performance in terms of resource sharing, whereas ILP outperforms all of them by 6-16%. Due to the limited number of working paths in each PG, ILP can handle dynamically arrived connection requests in a reasonable amount of time. Also, we find that the number of affected working paths in GSP is about half of that in SSR. We conclude that GSP provides a scalable and efficient solution for dynamic spare capacity reconfiguration following the (M:N)n control architecture.