The Asia-Pacific telecom sector is growing fast, and will soon have to come to terms with needing to do more with less, in particular to ensure precious bandwidth resources are utilized to their fullest for an increasingly data-hungry customer base. New techniques like optical spectrum mining provide an agile approach to unshackle market growth from current technological and operational constraints, as well as boosting revenues and customer experience alike.
No one is in any doubt that the APAC telecoms sector is growing, and growing fast. This brings about not only change and opportunity, but also enormous industry challenges.
In 2016, Asia already accounted for the largest share of the global mobile data traffic at 3.6 Exabytes per month, double that of North America. According to Ericsson’s November 2016 Mobility Report, mobile data traffic growth in Asia is expected to grow 8x by 2022.
The rise in data traffic is due not only to rising mobile penetration, but also what APAC consumers are doing with their mobile devices. Video content already accounts for a significant amount of mobile data consumed in the region, and as more people choose to stream content from their smartphone, networks will need to take on even more data loads.
In addition, the continuing expansion of connected things, a trend more widely known as the Internet of Things (IoT), that ranges from self-driving smart cars to remote sensors to smart cities is also adding pressure to the network. These trends require local service providers to enhance their network in support of the increasing number of internet-enabled devices in use at any given time across Asia.
This rapid rise of data traffic is leading to an important shift toward extracting more value from networks. 4G is already promising to double spectrum utilization, and increased densification – devices per unit area – will also play a considerable role. In this new digital era, rigid and static optical networks are no longer up to the task of economically addressing APAC’s rising demand for more and more data.
There is a new mandate to make the optical wireline spectrum work at maximum efficiency – in other words, doing a lot more with only what you have now.
Enter optical spectrum mining
To understand the importance of optical spectrum mining, it is key to consider the present mode of operation and how the network is designed.
Optical networks are largely deployed as rigid, static networks engineered for worst-case scenarios, including reduced efficiency due to system aging. This means that the actual capacity of the network is not fully utilized, because the network is typically designed with the assumption that all available end of life (EOL) margin should be reserved from day one. Too often, networks are built using best-guess predictions of worst-case conditions, making much of the available infrastructure inflexible in the face of changing needs.
Looking at future modes of operation, we see a need for more emphasis on identifying available margin and making it available when and where it’s needed to dynamically boost capacity on a temporary basis. Optical spectrum mining allows network operators to quickly increase revenue streams by making better use of the existing optical network capacity available. Coping with surges in demand will no longer mean costly and time-consuming manual intervention, or bringing new hardware online. Instead, available optical spectrum can be identified and made available rapidly on a temporary basis.
Spectrum mining can be brought to fruition by combining programmable hardware with advanced software capabilities. The following building blocks are essential:
- Programmable capacity: To extract the maximum value from existing network assets, operators must deploy flexible transceivers with variable bit-rate, software-configurable coherent optics that can match optimal channel capacity to available system margins for a specific network path.
- Fully reconfigurable: Optical spectrum mining requires true end-to-end visibility and programmability of all available resources to be able to reroute channels of variable spectral occupancy across any path, and across any optical spectrum in the network.
- Adaptive net system margin: To facilitate the reallocation of optical spectrum, operators need the ability to monitor, control, and extract the maximum value from their existing network assets at any point in time
- Instrumentation and control: In any software-defined networking model, there will be a constant need for many different systems to work in concert. Optical spectrum mining software applications abstract complexity associated with advanced flexible technologies, enabling operators to fully operationalize and realize benefits associated with the modernized network.
- Automation, analytics, and software defined networking: New control technologies like SDN, coupled with high degrees of instrumentation and analytics enable a feedback loop, where a network can essentially measure and adjust itself as per operator policies. These can be implemented as event driven policies, or real-time policies that run continuously. Innovative applications can be deployed on top of the network to ensure the most control over deployment of precious bandwidth assets.
With these building blocks in place, optical spectrum mining can help operators successfully handle temporary demands for higher bandwidth, that could relate to major events such as new data center commissioning, or database backup activities.
Optical spectrum mining heralds a new era for fault recovery, as agile transceivers can be used to bypass faults, such as fiber breaks to avoid long-term service outages instead of having to build a significant amount of redundant capacity into a network. In the Indian market, for example, it is estimated that as much as a third of all costs faced by network operators are caused by resiliency issues.
Optical spectrum mining is significant to the evolution of cloud-based consumption because it redefines optical networks for the on-demand world. It changes the operational paradigm for these networks, moving them from the static, “one-size-fits-all, set-and-forget” model to open and programmable assets that can deliver the exact service performance required where and when it is needed. Thus, operators can tune, control, and dynamically adjust optical network capacity as needed to more efficiently address constantly changing user demands, ultimately helping them meet the networking challenges of the next decade.
Written by Karl Horne, Chief Technology Officer of APAC at Ciena
Be the first to comment