Abstract

The traffic in metro and core networks is forecasted to grow in volume but also in dynamicity. Network operators dimension their optical networks and their IP edges for the expected traffic peak, and also reserve additional resources for the required survivability level. The latter is typically done by protection mechanisms at the optical or IP layer only. Multi-layer proactive restoration techniques can reduce the cost by enabling resource sharing while providing the same level of survivability. In this work we formulate the multi-layer survivability problem for an IP-over-elastic-optical-network and present ILP formulations to solve it, targeting two survivability levels: (i) single optical link and (ii) single optical link or optical and IP node failure. The proposed ILP algorithms are also distinguished with respect to their failure consideration point: (i) sequential, where we assume that the network is dimensioned for normal operation and then re-dimensioned to be resilient, and (ii) joint, where the network is dimensioned in a single step, considering normal and all single failure operation. We exploit the proactive (a priori provisioning) restoration concept to achieve sharing of the backup resources among different failure states. The proposed multi-layer techniques enable even higher efficiency, exploiting the IP grooming capabilities to enable the sharing of backup resources for a specific failure state but also the sharing of primary and backup resources. Compared to traditional single-layer protection approaches, the proposed joint multi-layer techniques were shown to yield significant cost savings.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call