Power-aware design of the optical interconnect for future data centers

The volume of traffic carried on the Optical Transport Network (OTN) is continuously growing, not only due to the increase of new services, new applications, and the number of both connected users and smart devices, but also owing to the rapidly growing Data Center Interconnect (DCI) market. As a result, telecom operators need to upgrade their transport networks to cope with these new traffic requirements. This tremendous increase in bandwidth demand will also introduce energy bottlenecks in OTNs. The telecommunication networks' significant Greenhouse Gas Emissions (GGE) is another challenge that must be addressed in OTN design and planning policies. Thus, energy-efficient OTN architectures are currently attracting the attention of telecom operators. The candidate OTN architectures are: (1) the conventional Dense Wavelength Division Multiplexing (DWDM) fixed grid with high Single Line Rates (SLRs); (2) the conventional DWDM fixed grid adopting Mixed Line Rates (MLRs); and (3) the Elastic Optical Network (EON) or flexigrid-based architecture which is enabled by the Optical Orthogonal Frequency Division Multiplexing (O-OFDM) technique. This paper evaluates the power consumption of the abovementioned OTN architectures under different traffic loads and patterns, as well as different network physical topologies. In order to perform the evaluation, new heuristic algorithms are specifically developed for the green design and planning of data center optical interconnects.

[1]  Biswanath Mukherjee,et al.  Spectrum management in heterogeneous bandwidth optical networks , 2014, Opt. Switch. Netw..

[2]  Francesco Musumeci,et al.  Protection in optical transport networks with fixed and flexible grid: Cost and energy efficiency evaluation , 2014, Opt. Switch. Netw..

[3]  Zuqing Zhu,et al.  Spectrum-efficient anycast in elastic optical inter-datacenter networks , 2014, Opt. Switch. Netw..

[4]  Jawad A. Salehi,et al.  Energy-Efficient Manycast Routing and Spectrum Assignment in Elastic Optical Networks for Cloud Computing Environment , 2015, Journal of Lightwave Technology.

[5]  Suresh Singh,et al.  Greening of the internet , 2003, SIGCOMM '03.

[6]  A. Macho,et al.  Optimized amplifier placements for improved energy and spectral efficiency in protected mixed-line-rate networks , 2014, OFC 2014.

[7]  Gangxiang Shen,et al.  Energy-Minimized Design for IP Over WDM Networks , 2012, IEEE/OSA Journal of Optical Communications and Networking.

[8]  V. Lopez,et al.  Traffic and power-aware protection scheme in Elastic Optical Networks , 2012, 2012 15th International Telecommunications Network Strategy and Planning Symposium (NETWORKS).

[9]  Annalisa Morea,et al.  Cost-efficiency of mixed 10-40-100Gb/s networks and elastic optical networks , 2011, 2011 Optical Fiber Communication Conference and Exposition and the National Fiber Optic Engineers Conference.

[10]  R.S. Tucker,et al.  Photonic Switching and the Energy Bottleneck , 2007, 2007 Photonics in Switching.

[11]  Ting Wang,et al.  On the Design of Energy-Efficient Mixed-Line-Rate (MLR) Optical Networks , 2012, Journal of Lightwave Technology.

[12]  Ulrich Gebhard,et al.  Performance of elastic and mixed-line-rate scenarios for a real IP over DWDM network with more than 1000 nodes [invited] , 2013, IEEE/OSA Journal of Optical Communications and Networking.

[13]  Imrich Chlamtac,et al.  Lightpath communications: an approach to high bandwidth optical WAN's , 1992, IEEE Trans. Commun..

[14]  Kumar N. Sivarajan,et al.  Design of Logical Topologies for Wavelength-Routed Optical Networks , 1996, IEEE J. Sel. Areas Commun..

[15]  Gangxiang Shen,et al.  Energy-minimized design for IP over WDM networks under modular router line cards , 2009, 2012 1st IEEE International Conference on Communications in China (ICCC).

[16]  Edsger W. Dijkstra,et al.  A note on two problems in connexion with graphs , 1959, Numerische Mathematik.