Stochastic hierarchical multiobjective routing model in MPLS networks with two service classes: an experimental study on imprecision and uncertainty issues

Abstract

The paper begins by reviewing a two-level hierarchical multicriteria routing model for Multiprotocol Label Switching networks with two service classes (Quality of Service and Best Effort services) and alternative routing, previously proposed by the authors. The features of the considered resolution heuristic are described. Some key issues raised by its complexity are discussed, as well as the major factors that constitute the sources of imprecision, inaccuracy, and uncertainty of the model and the way in which they are dealt with in the adopted resolution approach. Analytic and stochastic discrete-event simulation experiments are performed for different test networks, including experiments with a dynamic version of the routing method. This case study allows for the evaluation of the inaccuracies intrinsic to the analytic/numerical resolution procedures and of the uncertainty associated with the estimates of the mean of the stochastic traffic flows. An analysis focused on key robustness aspects of the model is also carried out.

Hierarchical Multiobjective Routing in MPLS Networks with Two Service Classes – A Meta-Heuristic Solution

“…The study of the heuristic approach HMOR-S2, the basis of which was reviewed in the previous section, was completed with a sensitivity analysis, which led to the consideration of variants of this heuristic. In the report [21], two variants to the HMOR-S2 were described, firstly the HMOR-S2 R where a floating relaxation was imposed on one of the first level objective function values, and secondly the HMOR-S2 B where a floating barrier was imposed on one of the first level objective function values. Extensive experimental analysis was carried out for those variants and a simulation study was also conducted.…”

“…This means that the HMOR-S2 heuristic involves a significantly lower number of calculations than HMOR-S2 SA . For further details on these calculations, see [35]. These complexity measures are an indication of the heuristic numerical complexity just at the level of the "optimization" procedures.…”

“…The complete formalization of the TS meta-heuristic version of HMOR-S2, HMOR-S2 TS , is in Appendix B.2 of the report [35]. As for the numerical complexity of this heuristic, the instructions in the inner cycle of the procedure are executed C Therefore, the numerical complexity represented by any of these measures is the same as for the HMOR-S2 heuristic (see [21]). For further details on these calculations, see also [35].…”

“…2. The complete formalization of the meta-heuristic version of HMOR-S2 using SA, HMOR-S2 SA , is in Appendix B.1 of the report [35]. Note that one of the features of a standard SA technique is the random choice of the new solution (to be taken into account at each step of the algorithm) among all the feasible solutions in the neighbourhood of the best solution found so far.…”