There is tremendous opportunity today for carriers willing to implement a more nimble network infrastructure, one able to support revenue-generating services on-demand. Legacy networks are too inflexible. Activating new services in these environments is a time-consuming and manual process, and incorporating the latest generation of dynamic services is almost impossible.
Many carriers are coming to realize that converging traditional IT with network operations creates a far more responsive infrastructure. Resources on which services depend can be programmable and fully elastic, i.e. cloud-like, instead of static and hardware-restricted. Services can even be delivered like applications through automated activation with customer control and visibility.
Carriers that embrace this change have the opportunity to evolve network connectivity business models to network-as-a-service (NaaS) platforms. As a result, they can better serve existing customers and open up markets by providing high value services to over-the-top service (OTT) providers, turning competitors to customers.
Revenue Generating Services
Advancing software-based networking technologies, namely Software Defined Networking (SDN) and Network Function Virtualization (NFV), virtualize the service-delivering functions of the network. By virtualizing – or abstracting – what was a previously static network infrastructure, new revenue-generating services can be turned up without the need to first manually allocate supporting IT resources. Instead, applications are able to request network characteristics, such as routing instructions and optimal bandwidth, on demand.
As a result, not only can new services be added far more quickly, but those services are automatically provided with the infrastructure support required for optimal performance. Optical networks using SDN control play a critical role in creating and dynamically configuring infrastructure that allows services to access network connectivity that is truly guaranteed, uncongested and high speed. This stands in stark contrast to the traditional “best effort” networking approach in which all services are allocated the same resources regardless of the type of traffic being transported.
For example, applications that depend on cloud-bursting (moving compute jobs across cloud boundaries) can trigger dynamic high-speed network connectivity to support the short-term large data transfer that often occurs, only for the extent of time or data volume that is required.
With SDN, NFV and open orchestration technologies, carriers can offer NaaS capabilities to each other. This can improve the way off-net or roaming services are provided across the globe. Rather than paying an off-network (or out of region) provider to offer its version of a given service, or to simply provide long distance connectivity back to the service instance on a trust-based billing arrangement, a service provider could instantiate its own service image in another carrier’s network. This can ensure a consistent service implementation and behavior regardless of where the originating service provider’s customers travel. This concept can be extended beyond simple voice and data roaming to include a broad range of content-based services.
Traditional Network Architecture Challenges
While current technology trends offer a bright future, it is not without challenges. Operational environments in traditional networks are often focused on specific network layers, with staff and resources divided between transport and Internet Protocol (IP) networks. While this method of organization is historically very common, it creates barriers between networking components and impedes the development of a holistic, dynamic infrastructure.
This legacy architecture is highly restrictive in many ways, for example: the available network and service functionality is typically locked in by proprietary hardware; IP-based functionality only peers with other IP resources; routing protocols inherently only run in their layer and there is no easy way to make changes in other layers; the operational support system (OSS) and network management system (NMS) are focused on specific layers of the network without dynamic interworking; and the operations departments themselves are siloed with separate staff and skillsets divided between network layers.
To date, these restrictive architectures have had at their core static hardware that encapsulates significant amounts of network and service functionality. The high cost and time required to implement vendor proprietary systems designed for a specific type of service means only certain groups of users at specific locations will have access to those new services. Understandably, carriers will only roll out these services where the necessary investment can justify the risk and expected uptake. As a result, service functionality is often limited to the “common denominator” between proprietary systems for full customer access.
The Solution to More Dynamic Carrier Networks
The integration of software control via SDN, software-based functions via NFV, and tools for service orchestration are the technological keys for unlocking dynamic capabilities in carrier networks.
As carriers take the first steps toward overcoming the challenges of replacing legacy systems, initial deployments of SDN and NFV are demonstrating the value of converging the previously disparate worlds of IT and transport networks. These technologies facilitate the disaggregation of traditionally complex, vendor-embedded functionality into open, dynamic platforms connected by de facto software interfaces. Rather than being dependent on hardware-embedded functionality, interoperability (including between network layers) evolves into a capability supported by a software-only layer.
Open standards and software platforms are driving this evolution. Previously, standards in different layers did not provide for configuration of, and interoperability across, layers of the network. However, the open software-based initiatives, such as OpenFlow and NetConf, now start to allow dynamic multi-layer networking to be a reality. Industry initiatives, OpenStack, OpenDaylight and ONOS for example, deliver the open software platforms and interfaces necessary to break open the vendor-embedded network hardware.
These initiatives are facilitating the development of new business models and open partnership arrangements among a broad range of contributing organizations.
The Future of Carrier Networking
The growing interest in, and commitment to, enabling more dynamic infrastructures means services will be introduced at faster rates and to a greater number of consumers. Having said that, it’s also important not to get caught up in the hype. The changes underway are real, but they’re not instantaneous. Over the course of the next 12-18 months, most carrier deployments will continue to be based on proof-of-concept trials. Technology vendors capable of integrating a broad range of components into working solutions will lead the activities.
We will see the increasing commercial availability of software and hardware products needed to run this dynamic environment with the stability, resiliency and reliability necessary for telecommunications: SDN controllers, NFV infrastructure, service apps in the form of virtualized network functions and orchestration platforms. Some of this will be in the form of open source software, but all software products, whether proprietary or open source, will need to openly interoperate with others. Vendors with strong networking and IT capability will play a pivotal role in helping all carriers evolve their network and service platforms. Competitors who traditionally created closed, proprietary solutions will need to learn to cooperate in open commercial and technical environments.
As the momentum builds, we will find ourselves working in a world where services are no longer restricted to a centralized cloud infrastructure. Accompanying the decrease in centralization will be the greater “IT-ification” of the network, utilizing more broadly distributed computing and storage resources; this will need to be orchestrated synchronously with the proliferation of network connected devices, which are expected to surge with the popularity of the Internet of Things (IoT).
Based on this dynamically available and open infrastructure, the cloud will extend and distribute itself leading to the evolution of “fog computing.” When combined with the evolution of social computing, mobility, cloud, big data, and the IoT, the result may be what IDC calls the“3rdPlatform.” Central to the 3rd Platform is a dynamic, open, multi-layer communications network. This network will be “Human Centric,” harnessing the creativity of people where it can be directed toward positive social outcomes and greater sustainability.
Automating so much of what has been expensive, time-consuming, manual work will pay dividends for carriers in the form of new revenue streams and reduced churn, thanks to a more satisfied customer base. As with all significant technology innovations, this new era of dynamic networking will have its share of “irrational exuberance.” Should hardware and software providers continue to deliver performance reliability, and to demonstrate the economic advantages of virtualized networks, operators will be able to discern what’s real from what’s mostly smoke and mirrors. A network transition of this magnitude is not to be entered into lightly, but when done right, opens up a world of opportunity.