Today’s telecommunications networks can be viewed as sprawling, service-specific technology quilts, patched together over decades of expansion and equipment upgrades. They feature multiple layers, with a set of statically connected network functions supporting service delivery, provisioning, and operations. Each site requires critical infrastructure systems including power, backup power, and thermal management; however, management and monitoring of these systems has been primarily a manual function limited to alarms when something failed. Keeping equipment supplied with conditioned energy and maintaining proper operating temperatures are critical functions for telecom network operation, and they also have been among the major contributors to increasing opex and capex costs.

Unprecedented global mobile data traffic growth over the last few years and pressures on network capacity driven by video applications, smartphones, and tablets are propelling the transformation of these telecom networks. Technological innovations – specifically software-defined networking and network functions virtualization – are fundamentally changing our approach to next-generation networks. The evolution to software-centric virtualized networks is taking hold on a global basis, delivering numerous benefits to providers, including new revenue streams, greater network agility, significant reduction in time to market, and substantial cost benefits.  

Complicating matters – or simplifying them, depending on your point of view – the onset of NFV and SDN separates network services and loads from dedicated hardware and servers, enabling a dynamic, virtualized network that runs on commercial-off-the-shelf hardware, transforming telecom network sites to IT look-alikes.

However, to realize the agility gains with more programmable, virtualized next-generation telecom network infrastructure and monetize the cost savings, providers need to move quickly to automate manual processes, collect, and analyze data across key network devices – including critical infrastructure – and scale resources to meet demand.

Imagine the potential of a truly software-centric virtualized telecom network, with the ability to shift loads seamlessly across devices and locations, from central office to remote sites at the outer reaches of the network. Equipment faults are addressed through capacity management, and monitoring is software-based. It’s truly transformative, making the next-generation communications networks more dynamic in every way, capable of adjusting not only to planned peaks like those around the Super Bowl, but to unanticipated spikes often triggered by major crises. That’s where we’re headed, but understand this: Shifting loads across virtualized networks to navigate peaks and spikes and ensure optimal operation and delivery of service is only possible if the power and cooling infrastructure is capable of supporting that load.

Of course, this kind of real-time managed software-centric and virtualized telecom network requires a critical network infrastructure that’s just as agile, flexible, and dynamic, with visibility down to power and thermal system devices in real time. It also requires vendor-agnostic critical infrastructure management systems that provide a single, consolidated view of these devices. These systems must navigate the underlying complexity of power and cooling hardware, aggregating, normalizing, and analyzing operational data from these devices. These management platforms are small analytical engines of their own, automating functions and delivering smart operational information and even predictive capabilities for use in control, orchestration, management, and policy layers of telecom networks, all in real time.

That last point is important. Management of critical infrastructure in real time ensures optimized performance – including reduced energy consumption and improved overall operating efficiency – of this increasingly dynamic software-centric virtualized telecom network.

So, we’re on the precipice of a new network, managed through a single pane of glass, with visibility into power systems and cooling units, and the ability to activate and adjust those systems in real time across all layers of the network. This vendor-agnostic management platform will see everything, communicate with each piece of equipment using a common language, analyze data, and predict potential equipment failures. Expensive maintenance calls will become efficient, proactive service calls, addressing potential problems before they happen with exactly the personnel, parts, and tools needed. Operational expenses will shrink as the network becomes more efficient and those truck rolls more purposeful.

A decade ago, this would have sounded like science fiction. Today it’s happening. And it’s only the beginning.

Henry Towster is vice president of business development, strategic accounts of Emerson Network Power.