Survivability analysis of restorable SONET networks.

摘要

For SONET mesh networks, the restoration process plays an important role in determining the survivability. Thus, survivability evaluation of the network, with the ability to quantify the differences between restoration techniques, has become essential. In this dissertation, we first introduce a model called parametric State Reward Markov Model, SRMM/p, and propose a set of metrics in order to evaluate the survivability and performability of the restorable SONET mesh networks. We incorporate the performance level of the network into these metrics in the form of rewards that are calculated globally with topology-free approximations. Even though the metrics are closely related, they capture different aspects of the networks and could be used in a complementary way in design and deployment decisions. Further, we present a solution to the high run-time complexity of the model which is caused by the disparity between transition rates amongst various states in the SRMM/p. The solution uses the state aggregation approach and is validated by an experimental study.;The SRMM/p is equally applicable to the analysis of self-healing SONET rings by considering a ring as a special case of a mesh topology. Further, the topological uniformity and simplicity of the rings allow one to include more detailed features of the network in the model so that the analysis will be more accurate. For this purpose, we propose an improved approach to the survivability analysis of self-healing SONET rings which employs a probability-tree based evaluation of the probability of various system states resulting from all possible combinations of node and link failures. Survivability of rings with distinctive demand patterns is studied with the improved analysis and compared experimentally.;Moreover, by recognizing the incremental expansion of SONET network systems, we propose a hierarchical approach to survivability and performability analysis of large hybrid restorable SONET networks. The sensitivity analysis of the survivability and performability figures of a large hybrid SONET system is carried out through a set of experiments. As the final product of the dissertation, we introduce a stand alone easy-to-use tool namely SONSAT (SONET S&barbelow;urvivability A&barbelow;nalysis T&barbelow;ool).