Reduced footprint and power consumption: Estimations on the ­power consumption of pure photonic switching devices is – even in worst case scenario – still one order of magnitude lower than traditional electronic telecom network core equipment. This also includes the footprint of the equipment. It scales down the number of units for supplemental equipment as well (uninterruptible power supplies, air condition etc.). Savings in terms of OPEX are obvious and confirm the overall trends with environmental friendly, low-power consuming Green ICT solutions.   

Enhanced revenue streams: The benefit of networks based on an intelligent control plane extends beyond survivability and ­lower OPEX/CAPEX. It enhances revenue streams directly as it ­enables real-time inventory management and supports the ­delivery of services on demand. New services can be directly ­created. The next generation of optical networks can handle ­larger pool of diversified services, hence can support more clients.  

Figure 3 compares footprint [normalized to Cisco racks], power consumption [kW] and switching capacity [Tbps] of today’s available OEO routers (Cisco CRS-1 and Juniper T640) with the OOO approach (Photonic Packet Switching). Given the fact that PPS equipment is not available on the market yet, the values for power consumption for OEO are reduced by a factor of 3 (allowing room for improvement) whereas for OOO the power requirement is increased threefold (allowing large errors often present in R&D).² Even in the face of these corrections (almost tenfold), the OOO approach outpaces its electronic OEO competitor by more than a factor of 10 in foot print and power requirements.

Operators should take on the challenge and design their “Next Generation” optical core network architecture to easily include/extend it with PPS technology, combined with wavelength routing functions to ensure a highly efficient, reliable, and dynamic network.    

Although PPS offers much higher capacity than today’s electronic packet switching, it will not completely take over and networks will not evolve within the next week into an “all-optical” network. Enhanced hybrid versions (operating on the electronic and optical level) will come first. The electronic domain is (still) required to perform BER, QoS, and protection and ­restoration functions in the optical network. It is still challenging and ­costly to achieve functions such as performance monitoring, ­sub-lambda grooming, advanced and flexible protection schemes, and dynamic line equalization on a photonics level. Ironically, it will be the current superior “intelligence” of electronics that will further support efficient and fast implementation of PPS.