Network Computing is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

'Cloudified' Telecom Services: What To Look For

Mobile and broadband communications service providers find themselves in a challenging position. Network usage is growing rapidly -- 50% or more annually -- as subscribers consume huge quantities of bandwidth-intensive services from Google, Amazon, Netflix, Hulu and other content providers. Meanwhile, revenues are relatively flat, as content providers cash in at the carriers' expense.

Simply adding more bandwidth using traditional capex strategies is not sustainable. Existing network infrastructure designs are too costly and inefficient. What's needed is a paradigm shift that drives down capex and opex, while opening the door to the next-generation services that will keep carriers from becoming unprofitable commodity players.

How will carriers overcome this challenge? By moving to virtualized, software-defined and "cloudified" networks that transform not only carriers' cost structures but also their business models -- and their competitive stance. But how will they get from here to there? And where is "there" anyway? Like any major paradigm shift, this transformation will be incremental. The four advances below will lead to "cloudified" telecom networks that can support the though demands of modern enterprise customers. 

1. Network functions virtualization

Virtualization has transformed the efficiency and flexibility of enterprise networks. Now, it's the telecom provider's turn -- and Network Functions Virtualization (NFV) is the first step in this direction. Instead of siloed, proprietary systems dedicated to specific network and service functions, these functions are performed in software on COTS hardware. In essence, this provides a "liquid pool" of dynamically allocated resources able to perform any function with policy-based control.

Many telecom providers are also deploying software-defined networks (SDN) to further abstract the network's underlying data plane from the control plane and making it aware of the underlying switching infrastructure. Instead of each network component having to individually identify its peers and what it should be doing in the presence of those peers, SDN enables efficient centralized routing, significantly reducing operating costs. 

This represents a revolutionary shift for telecom operators, dramatically driving down the cost of doing business while giving them a degree of agility they could only dream of with their rigid legacy architectures. This shift is already underway. A number of leading players are conducting trials of NFV implementations now, with plans to commence commercial deployments by 2016.

2. Virtualized resilience layer

Once operators have begun their virtualization journey through NFV, the next step is to migrate to cloud technologies that open the door to new revenue-generating opportunities. Adding sophisticated cloud orchestration capabilities enables operators to deliver innovative, new services that dynamically combine service elements and functions to create highly personalized subscriber experiences with self-provisioning. Sophisticated services once available only to large enterprise subscribers can now be delivered to SMEs or even individuals quickly and cost-effectively -- meaning profitably.

Cloud technologies also provide the automation required to enable massive scaling at very low cost. This means automating policy-based workload management across huge numbers of servers distributed across geographic regions. So operators can scale their subscriber bases and service offerings without dramatic increases in capex and opex. 

3. Contextual network analysis

The third step revolves around leveraging the information operators accumulate about subscribers -- devices, service plans, usage patterns, locations, purchasing history, personal and business contacts, and more. By combining this knowledge with insights from social media and other sources, operators could create massive repositories of information that could be used to deliver highly personalized opt-in ads, offers and recommendations. This offers the potential for third-party partnerships that enable profitable, new value-added services to attract new subscribers and to improve retention by making services stickier.

4. Thinking networks

The final step in telecom cloudification is taking the network to an even higher level of intelligence and automation. Here, the network has a high level of software-defined intelligence across all central office domains, creating a 360-degree view of the network and its subscribers. This would enable the network to adapt and respond automatically, learning as it goes to predict what's needed, when and where. The result is a "thinking network" that is able to deliver an optimized subscriber experience while optimizing network profits.

Clearing the availability hurdle

These steps describe an exciting vision for carriers and a path to a profitable future. But, as with any paradigm shift, they are easier said than done. One of the reasons telecom operators are is lagging in the shift to virtualization and cloudification has to do with its extreme five nines availability requirements. Subscriber applications must be able to maintain state -- "remembering" the preceding events in a given sequence of user interactions -- in the event of a fault or failure. Losing that state equals dropped calls, delayed messages, lost revenue, and subscriber churn.

Until now, achieving stateful reliability in large-scale, low-cost cloud environments has been challenging. But savvy telecom providers have discovered the solution in a new generation of software-defined availability technologies that capture the state of the primary system at regular intervals and apply it to a secondary standby host. In the event of a primary host fault, the secondary can pick up execution starting from the most recent statepoint without losing any data, completely transparent to the subscriber. 

Note that taking availability out of the application layer is key to success. So any application with any availability requirement can be run in the cloud with application transparency.

Thinking networks may be a way down the road. But taking those first steps today toward a software-defined future will be crucial for carriers and the services their customers use to build their enterprise infrastructures.