Abstract

In this paper, we investigate power-efficient directed preconfigured cycles ( $p$ -Cycles) for asymmetric traffic protection in elastic optical networks (EONs) against single link failure. Owing to the advantage of distinguishing traffic amount in two directions, directed $p$ -cycles consume low power by allocating different spectrum slots and modulation formats for each direction. A mixed integer linear programming (MILP) model is formulated to minimize total power consumption, which takes into account directed cycle generation, spectrum allocation, modulation adaptation, and protection capacity. To increase the scalability, the MILP model is decomposed, and a two-step approach is proposed: improved cycle enumeration and a simplified integer linear programming model. Extensive simulations are performed to study the power consumption of $p$ -cycles under different traffic patterns in terms of traffic asymmetry ( $TASY$ ), anycast ratio (AR), and the number of data centers (DCs). The results strongly demonstrate that directed $p$ -cycles obtain significant power savings for protecting asymmetric traffic in EONs. The power savings rise up to $46.91\%$ and $36.38\%$ compared with undirected $p$ -cycles as the $TASY$ and AR increase, respectively. Moreover, the directed $p$ -cycles achieve valuable power savings (up to $46.1\%$ ) with the introduction of DCs while the amount of power savings does not depend on the number of DCs.

© 2016 IEEE

PDF Article

References

You do not have subscription access to this journal. Citation lists with outbound citation links are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access OSA Member Subscription

Cited By

You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access OSA Member Subscription